Molecular Nature of 11 Spontaneous de Novo Mutations in Drosophila melanogaster

Corresponding author: Hsiao-Pei Yang, Department of Ecology and Evolutionary Biology, Corson Hall, Cornell University, Ithaca, NY 14853., hy31{at}cornell.edu (E-mail)

Communicating editor: D. CHARLESWORTH

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

To investigate the molecular nature and rate of spontaneous mutation in Drosophila melanogaster, we screened 887,000 individuals for de novo recessive loss-of-function mutations at eight loci that affect eye color. In total, 28 mutants were found in 16 independent events (13 singletons and three clusters). The molecular nature of the 13 events was analyzed. Coding exons of the locus were affected by insertions or deletions >100 nucleotides long (6 events), short frameshift insertions or deletions (4 events), and replacement nucleotide substitutions (1 event). In the case of 2 mutant alleles, coding regions were not affected. Because 70% of spontaneous de novo loss-of-function mutations in Homo sapiens are due to nucleotide substitutions within coding regions, insertions and deletions appear to play a much larger role in spontaneous mutation in D. melanogaster than in H. sapiens. If so, the per nucleotide mutation rate in D. melanogaster may be lower than in H. sapiens, even if their per locus mutation rates are similar.

SPONTANEOUS mutation is the key genetic process that supplies raw material for stabilizing (deleterious mutations) and directional (beneficial mutations) natural selection. However, we know relatively little about the molecular nature of this process, i.e., about how common different events (nucleotide substitutions, deletions, insertions, duplications, etc.) are among all spontaneous mutations, or about the value of its basic quantitative parameter, the per nucleotide per generation spontaneous mutation rate µ (DRAKE et al. 1998 ; KONDRASHOV 1998 ).

Two approaches can be used to estimate µ. First, one can measure the degree of divergence between homologous selectively neutral DNA sequences in related species, provided that the total number of generations from their last common ancestor is known with good precision. The best data of this kind are available for the human-chimpanzee pair, where sequence divergence, mostly due to nucleotide substitutions, between orthologous pseudogenes is 1.3%. This implies, assuming a 20-year generation time and 5 million years since the last common ancestor, that µ 2 x 10^-8 (NACHMAN and CROWELL 2000 ). Unfortunately, this approach cannot currently be applied to Drosophila because the numbers of generations that separate pairs of Drosophila species are known, at best, only within a factor of 3–5, due to large uncertainties regarding the absolute divergence time and the number of generations per year in nature. Furthermore, since Drosophila populations are large, even weak selection can be effective, leading to selection on silent sites causing substantial codon bias (AKASHI 1997 ). Thus, it is difficult to be sure that a particular DNA sequence within a Drosophila genome evolved at a rate equal to the mutation rate (PRITCHARD and SCHAEFFER 1997 ). In contrast, there is little doubt that mutations in hominoid pseudogenes were effectively neutral.

Second, µ can be estimated from the per locus mutation rate, m. If the mutational target at a locus, i.e., the number of nucleotides whose changes will lead to a phenotypically detectable mutation, is t, µ = . However, the size of the mutational target depends strongly on the molecular nature of mutation. If we consider loss-of-function mutations, t for insertions and deletions is close to the total number N of protein-coding nucleotides at a locus, since most of such events (except in-frame insertions and deletions) lead to malfunction of the affected proteins. In contrast, t for nucleotide substitutions may be closer to N/5, since only 5% of substitutions create an in-frame stop codon, and <25% of missense substitutions (and very rare synonymous substitutions) lead to total loss of function (MOHRENWEISER 1994 ; KONDRASHOV 1998 ). Thus, studying the molecular nature of mutation is important in its own right and is also essential for estimating µ.

In humans, the majority of loss-of-function de novo spontaneous mutations are substitutions (KRAWCZAK et al. 2000 ). Thus, m = 10^-5 for a locus with 2000 coding nucleotides implies µ 2 x 10^-8, in agreement with the rate of neutral evolution. Data of this kind, first obtained for the hemophilia B locus (SOMMER 1995 ), are now available for >20 human loci (KRAWCZAK et al. 2000 ).

In contrast, we have little data on the molecular nature of de novo spontaneous mutation in Drosophila melanogaster. YAMAGUCHI et al. 1994 found that, out of six loss-of-function (null) mutations at the Gpdh locus obtained in a mutation-accumulation experiment, five are P-element insertions and one is a long deletion. Loss-of-function alleles segregating in wild populations of D. melanogaster can be caused by both minor and large-scale events (NITASAKA et al. 1995 ; TEN HAVE et al. 1995 ); unfortunately these authors did not discriminate between substitutions and short indels. Data on molecular evolution imply that indels (mostly deletions) constitute 20% of all mutations in Drosophila, i.e., substantially more than in mammals (PETROV et al. 1996 ; PRITCHARD and SCHAEFFER 1997 ; PETROV and HARTL 1998 ). PETROV and HARTL 1998 found that deletions are much more common than insertions, while PRITCHARD and SCHAEFFER 1997 claim that they are about equally common.

Thus, although extensive data on m are available for many loci of D. melanogaster (SCHALET 1960 ; WOODRUFF et al. 1983 ), they cannot currently be used to estimate µ. Obviously, we need more data on the molecular nature of mutation in Drosophila. Here, we report the results of specific locus tests (SLTs) of several loci that affect eye color; four loci are autosomal [chromosome 2, purple (pr), cinnabar (cn), and brown (bw); chromosome 3, scarlet (st)] and four are X-linked [prune (pn), vermilion (v), garnet (g), and white (w)]. Sixteen independent de novo spontaneous mutational events were detected in wild-type D. melanogaster, and molecular analysis of 13 mutations was performed.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Flies and cultural conditions:
We sampled 100 mated females from a large wild population of D. melanogaster near Ithaca, NY. Flies were bred in 2.5 x 9.0-cm vials. Each vial contained 8 ml of food (1% agar, 0.1% propionic acid, 10% brewers yeast, and 10% glucose) seeded with a few grains of live baker's yeast. At least 150 flies can develop simultaneously in such a vial without a significant increase of mortality. Flies were kept under a 12/12 light/dark cycle, at 25° and 75% humidity. Generation one (G₁) parents (see below) were stored at 16°, and remained fertile for at least 45 days. CO₂ anesthesia was used.

Tester strain and balancer strain:
The tester strain, V, which is homozygous for the alleles pr¹, cn¹, bw¹, st¹, and kar¹, and the balancer strain, STV, which has balancers on both autosomes (SM1 and TM3) and is heterozygous for alleles pr¹, cn¹, bw¹, st¹, and kar¹, were both created by introducing mutant alleles paternally into the flies that originated from the same wild population. Large population sizes, at least 1000 flies, were maintained at all intermediate stages of the strain creation. As a result, both strains are vigorous, despite carrying several marker alleles.

Detection of eye color mutation by SLT:
Offspring of different wild-caught females (G₀) were crossed individually to unrelated G₀ flies, and produced sibships of G₁ flies (Fig 1). From each G₁ sibship, at least 30 females were mated, in groups of 10, with strain V males (10–12 per group). Two or 3 days later, each group of females was allowed to lay eggs for 4 hr in a vial. During this time, 100–200 eggs were laid. After this, the mothers were removed and stored at 16°. The G₂ offspring that emerged in these vials were screened for salient eye-color phenotypes due to mutations at the four autosomal loci (kar mutants could not be reliably detected) and at X-linked loci that affect eye color (only in males). Mosaic mutants were ignored.

View larger version (22K): In this window In a new window Download PPT slide   Figure 1. Mating scheme for one family. R denotes meiosis, which forms the boundary between successive generations. S denotes the occurrence of syngamy. P denotes "perigametic interval," the time around meiosis (from the last premeiotic DNA replication until the first DNA replication within the zygote). Fi and Mi denote females and males of the ith generation.

Since our design involves large G₁ sibships, we can distinguish de novo clusters of mutations from preexisting heterozygosity. If an original G₀ fly carried a heterozygous loss-of function allele at pr, cn, bw, st, or kar, 25% the G₂ offspring of her or his G₁ daughters should have a mutant phenotype. Seven such families, each with >100 G₂ flies with abnormal eye color found in more than one vial, were identified and removed from the analysis. These families were clearly different from the three de novo clusters that we have found (see below). If a single mutant was found in a vial, the G₁ females that laid eggs in it were allowed to lay eggs again to make sure that the singleton was not actually a small cluster. Thus, we effectively studied the mutation process in G₁, while screening G₂.

Identifying and isolating mutant alleles:
Mutants detected in G₂ during screening were mated individually with flies of the balancer strain, STV. Mendelian segregation within the offspring from these crosses determined the chromosome affected by a mutation and, for mutations at the four autosomal loci (pr, cn, bw, and st), identified the affected locus. Identity of the mutant locus was then confirmed by the appropriate complementation tests performed on offspring from G₂ x STV cross. If a G₂ fly was not a mutant at one of our four autosomal loci, only male mutants were further analyzed by the appropriate complementation tests with the following four X-linked loci: w, g, v, and pn.

To isolate an autosome that carried the de novo mutant allele, we identified, using the appropriate crosses, male offspring from the G₂ x STV cross that carry both SM1 and TM3 balancers and an autosome carrying a mutant allele affecting eye color only at the locus where a de novo mutation occurred. Only one such autosome was analyzed if the G₂ mutant was a male (because there is no crossing over in males). Five chromosomes were analyzed if the G₂ mutant was a female (in all such cases, the mutation occurred at bw, and bw is far away from pr and cn on chromosome 2), and the mutant allele different from the one present in strain V was regarded as the new one. All mutations discovered in fertile G₂ individuals were homozygous viable and were, after being isolated, kept as pure strains, with the sole exception of the homozygous lethal loss-of-function mutation at pr, which was kept heterozygous with a second balancer strain (CyO).

Molecular characterization of mutant alleles:
Mutations found in our screen can be due to molecular events at different scales. Thus, we applied sequentially to each mutant three different techniques.

1. Cytogenetic analysis: Ectopic exchanges between transposable elements (TEs) situated in heterozygous positions around a locus at which mutations are screened can lead to cytogenetically detectable events. Slides of squashed salivary glands from five third instar larvae per mutant chromosome were prepared, following ASHBURNER 1989 , and scored for the presence of deletions and duplications. In all cases, no major chromosomal alterations were found.

2. Southern blot analysis: Southern blotting was performed to detect mutations caused by transpositions and other relatively large events. Genomic DNA of 50 flies of each mutant line was extracted with SDS lysis, phenol-chloroform extraction, and ethanol precipitation (SAMBROOK et al. 1989 ). About 1 µg genomic DNA was restricted with one or more restriction enzymes with 6-bp recognition sequences, size-fractionated in 1% agarose gels, and blotted onto S & S Nytran neutral charge membranes by capillary transfer (SAMBROOK et al. 1989 ). The membranes were probed with ³²P-random-primer-labeled DNA clones from the corresponding gene at 58° overnight in hybridization solution (0.01 g/ml BSA, 1 mM EDTA pH 7.2, 0.5 M Na₂HPO₄ pH 7.2, and 7.5% SDS) and washed at 50° for six to eight times in washing solution (40 mM Na₂HPO₄ pH 7.2, 1 mM EDTA pH 7.2, and 1% SDS). The hybridized membranes were exposed to X-OMAT films at -80° for 3–14 days. Maps of restriction sites and probed regions of the genes analyzed are shown in Fig 2.

   
   

   View larger version (24K): In this window In a new window Download PPT slide   Figure 2. Molecular structures of genes at the eight loci. Boxes indicate exons, and the solid and open boxes indicate the coding and nontranslated regions, respectively. Sizes of these genes are relative to the scale on the top. For genes analyzed by Southern blots, including cn, bw, pn, and v, locations of restriction sites analyzed and the probed regions are also shown in the bottom (shaded boxes) of each gene. More detailed sequence information can be found in references: pr, KIM et al. 1996 ; cn, WARREN et al. 1996 ; bw, MARTIN-MORRIS et al. 1993 ; st, TEARLE et al. 1989 ; pn, TENG et al. 1991 ; v, SEARLES et al. 1990 ; g, LLOYD 1998 ; w, O'HARE et al. 1984 .

3. Sequencing: For those mutant alleles where no large differences in the lengths of restriction fragments were detected, their coding regions were PCR amplified and sequenced. Primers for PCR reactions were designed using PRIMER3. Standard PCR conditions (100 ng genomic DNA template, 0.5 µg of each primer, 200 µM dNTPs, 1.5 mM MgCl₂, 10 mM Tris-HCl pH 8.4, 50 mM KCl, 1 unit of Taq polymerase; Promega, Madison, WI) were used with thermal cycles: 95° for 3 min, followed by 35 cycles of 92° for 30 sec and 60° for 1 min, and ending with 72° for 7 min. Length of DNA amplified is within the range of 300–500 bp. The amplified DNA was purified with Ultrafree-MC centrifugal filters (Millipore, Bedford, MA) for sequencing. Both DNA strains were directly sequenced by autosequencing. The sequences were compared to wild-type alleles deposited in GenBank (accession nos. pn, Z12141; v, M34147; w, U64875; g, U31351; pr, U36232; cn, U56245; bw, M20630; and st, U39739) using basic BLAST search (version 2.0). Whenever a mismatch between the mutant sequence and the wild-type sequence was found, the gene region where the mismatch was located was PCR amplified and sequenced again to confirm the reality of the difference.

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

We screened 887,000 D. melanogaster for spontaneous de novo loss-of-function mutations that occurred in the female germ line at four autosomal loci (pr, cn, bw, and st) and four X-linked loci (pn, v, g, and w). In total, 28 mutants were found in 16 independent mutational events. Three events were clusters, of 10, 3, and 2 mutants, respectively. No mutations were found at g or w. Also, six X-linked eye color mutations at loci other than pn, v, g, or w were found. These were not analyzed because the loci involved are not yet known. Thirteen of the 16 independent mutations were isolated into pure strains and analyzed molecularly. In 11 cases, the probable cause of the mutant phenotype was found (Table 1, Fig 3).

View larger version (184K): In this window In a new window Download PPT slide   Figure 3. Southern blot analysis of bw mutant alleles. Samples of genomic DNA from each strain were digested with EcoRI (a), and with EcoRI and HindIII (b). Probed fragments and their expected sizes in wild-type (wt) alleles are indicated. (Lanes 1–7 in a correspond to wt, B1, B2, B3, B4, B5, and bw1 and lanes 1–6 in b correspond to wt B5, B4, B1, B2, B3, and bw1 respectively. Thin lines on the left are marker sizes.)

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Dealing with clusters of mutations:
In multicellular organisms, a mutation can occur during any of many cellular generations that constitute a single organismal generation. As a result, a mutation can lead to a cluster of mutants (if it occurred well before gametogenesis within a parent of the individual screened), a singleton, or a mosaic (postzygotic) mutant (Fig 1; RUSSELL and RUSSELL 1996 ; THOMPSON et al. 1998 ; RUSSELL 1999 ).

Dealing with singletons is straightforward. Mosaics are hard to use for quantitative studies because they may be cryptic or difficult to detect (SCHALET 1960 , SCHALET 1986 ). Eye-color mosaics, in particular, are manifested phenotypically only for loci that have autonomous expression (bw and w among our eight loci) and only if a sizeable fraction of the eye surface happens to be affected. Thus, we ignored mosaics and concentrated on mutations that occurred during G₁ (WOODRUFF and THOMPSON 1992 ; DRAKE et al. 1998 ).

Any SLT must be designed in such a way that a cluster of de novo mutations can reliably be distinguished from preexisting heterozygosity. This is easy if a cluster is small, provided that enough offspring from each G₁ individual are analyzed. However, mutations occurring just before the perigametic interval, which is the time between the last premeiotic DNA replication and the first DNA replication within the zygote (P in Fig 1), which are quite common (RUSSELL and RUSSELL 1996 ; RUSSELL 1999 ), produce large clusters, such that up to 50% of gametes produced by the affected G₁ individual carry a mutation.

If such large clusters of de novo mutations are misclassified as preexisting heterozygotes, m can be underestimated by as much as a factor of 5 (SELBY 1998A , SELBY 1998B ). Fortunately, even large clusters of de novo mutations can easily be distinguished from preexisting heterozygosity if sibships of 20–30 or more G₁ individuals are analyzed. Thus, it is surprising that our SLT appears to be the first one to use this sibship-based design.

Molecular nature of mutations:
Among the 13 events analyzed, 5 mutations, all at bw, were insertions of lengths from 3100 to 4500 nucleotides, 1 was a deletion of 2000 nucleotides, 4 were short frameshift insertions or deletions in coding exons, and only 1 event, a cluster of 10 cn mutants, was probably due to a replacement nucleotide substitution (Table 1). Five long insertions were probably caused by transposable elements, although this is not certain.

The frequency of insertions and deletions among loss-of-function mutations must be 5–10 times higher than among all de novo mutations, because almost 100% of insertions and deletions, but only 10–20% of substitutions, within the coding regions inactivate a protein (KONDRASHOV 1998 ). Thus, our data imply that insertions and deletions, both major and short, constitute at least 20% of all spontaneous mutations in Drosophila, which agrees well with estimates by PETROV and HARTL 1998 ; PETROV and HARTL 1999 based on patterns of DNA sequence evolution.

Among loss-of-function mutations segregating in wild D. melanogaster populations, insertions and deletions are also common (NITASAKA et al. 1995 ; TEN HAVE et al. 1995 ). However, data on autosomal recessive loss-of-function alleles, both in Drosophila and in Homo, may not accurately reflect the properties of de novo mutation because many generations elapse between the origin of a mutation and its detection, which allows selection to act on heterozygotes.

Our data also agree with early results by MUKAI and COCKERHAM 1977 , VOELKER et al. 1980 , and HARADA et al. 1993 , who found that the mutation rate toward loss-of-function (null) alleles of several proteins is 10 times higher than toward active alleles with changed electrophoretic mobility. This is to be expected if insertions and deletions are common. Indeed, all six null mutants found at the Gpdh locus during a mutation-accumulation experiment were major insertions (five) or deletions (one; YAMAGUCHI et al. 1994 ).

The molecular nature of spontaneous mutation in humans is rather different. Replacement and nonsense nucleotide substitutions in coding regions cause 70% of X-linked recessive (SOMMER 1995 ; GIANNELLI et al. 1998 ; LEMAHIEU et al. 1999 ; TUFFERY-GIRAUD et al. 1999 ) and autosomal dominant (haplo-insufficient) de novo loss-of-function mutations (KRANTZ et al. 1998 ; PROSSER and VAN HEYNINGEN 1998 ; CLOUGH et al. 1999 ; LOHMANN 1999 ; MAYER et al. 1999 ; NIIDA et al. 1999 ; WESTERMAN et al. 1999 ; KRAWCZAK et al. 2000 ). For individual loci, however, the fraction of substitutions varies from >90% (hemophilia B, SOMMER 1995 ; GIANNELLI et al. 1998 ) to <50% (Alagille syndrome, KRANTZ et al. 1998 ). Thus, insertions and deletions should account for only 5% of all spontaneous human mutations, which is confirmed by data on DNA sequence evolution (NACHMAN and CROWELL 2000 ).

This difference may be at least partially due to methylation of mammalian DNA at CpG dinucleotides, which drastically increases substitution rates at many such nucleotides. As a result, >30% of human nucleotide substitutions occur at CpG hotspots (SHIANG et al. 1994 ; SOMMER 1995 ). In contrast, DNA in Drosophila is not methylated (see LYKO et al. 1999 ), and CpG dinucleotides apparently mutate at a normal rate.

Per locus mutation rate m:
Because mutations at the four X-linked loci could be detected only in male offspring, our estimate of the per locus rate for independent mutational events is k = = 3.0 x 10^-6, where B is the total number of loci screened. Since B = 4 x 887,000 + 4 x 443,500 = 5,322,000, and the total number of mutants found was 28 (Table 1), our estimate of the per locus mutation rate is m = = 5.3 x 10^-6. The number of rare independent events has a Poisson distribution, so that the 95% confidence interval for k is 1.7–4.9 x 10^-6. The confidence limits for m cannot be calculated precisely, due to insufficient data on the fraction of clusters among all mutational events and on the distribution of cluster size. Assuming that between 50 and 75% of all spontaneous mutations occur in clusters (THOMPSON et al. 1998 ), we can therefore tentatively conclude that 2.0 x 10^-6 < m < 15.0 x 10^-6.

This estimate is in agreement with those obtained previously for both D. melanogaster (3–5 x 10^-6; MUKAI and COCKERHAM 1977 ; VOELKER et al. 1980 ; WOODRUFF et al. 1983 ; HARADA et al. 1993 ) and mammals (DRAKE et al. 1998 ).

Per nucleotide mutation rate µ:
Despite similar values of m for loss-of-function mutations in Drosophila and in humans (DROST and LEE 1995 ), µ in Drosophila may be substantially smaller than in humans because the mutational target size for indels, which are more common in Drosophila, is 5–10 times larger than for substitutions.

Because the average length of the coding regions of the eight loci used in our screening is 1620 nucleotides, our estimate for the component of µ due to insertions and deletions is µ_indel = (1.2 10.0) x 10^-9. Not a single nonsense substitution was found in 887,000 flies screened at the four autosomal loci in which we calculate that there are 526 possible substitutions that would create an in-frame stop codon, nor in 443,500 flies screened at the four X-linked loci in which we calculate 775 possible substitutions that could create an in-frame stop codon. Thus, assuming that all nucleotide substitutions occur with the same rate, no nonsense substitutions happened within the target equivalent of = 2.7 x 10⁸ nucleotides (division over 3 is because a nucleotide can be substituted in three ways). If we ignore clustering, this implies that µ_sub < 1.2 x 10^-8 with 95% confidence.

In contrast to mammals, there is no evidence that the spontaneous mutation in Drosophila is male biased, and there are approximately equal numbers of cell divisions in the female and male germ lines (BAUER and AQUADRO 1997 ). Screening of 100,000 flies for mutations in the male germ line did not provide any evidence for elevated mutation rate (data not reported). Thus, we probably did not underestimate k and m by our procedure.

Out of 100 spontaneous de novo mutations that are estimated to occur in a human genome every generation, at least 2–3 are probably deleterious (EYRE-WALKER and KEIGHTLEY 1999 ). In contrast, the genomic deleterious mutation rate U in Drosophila is poorly known, although it is probably below that in humans (KONDRASHOV 1998 ). Drosophila, being sexual, may falsify the mutational deterministic hypothesis for the maintenance of sexual reproduction if it has U < 0.5–1.0 (KONDRASHOV 1988 ). Assuming that, as in Caenorhabditis elegans (SHABALINA and KONDRASHOV 1999 ), 30% of 3 x 10⁸ nucleotides in the D. melanogaster diploid genome are controlled by selection, we can conclude that U < 0.5–1.0 and, therefore, this hypothesis must be rejected for this species, if µ < 0.5–1.0 x 10^-8.

	ACKNOWLEDGMENTS

We thank S. A. Shabalina, F. A. Kondrashov and V. A. Kondrashov for helping with mutant screening; S. V. Nuzhdin for guidance in the molecular part of the work; R. MacIntyre and C. Webb for useful suggestions, the anonymous reviewer who suggested that CpG methylation in mammals may be responsible for differences between Drosophila and humans; and the following people for providing gene clones: W. Warren (cn), L. Searles (v), P. Kim (bw), and K. O'Hare (w). This work was supported by a Fellowship for Study Abroad from the Republic of China Government to H.-P. Yang and a National Science Foundation grant DEB-9815621 to Sergey V. Nuzhdin.

Manuscript received June 23, 2000; Accepted for publication November 27, 2000.

	LITERATURE CITED

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

AKASHI, H., 1997  Codon bias evolution in Drosophila: population genetics of mutation-selection drift. Gene 205:269-278[Medline].

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

BAUER, V. L. and C. F. AQUADRO, 1997  Rates of DNA sequence evolution are not sex-biased in Drosophila melanogaster and D. simulans.. Mol. Biol. Evol. 14:1252-1257[Abstract].

CLOUGH, M. V., J. D. HAMLINGTON, and I. MCINTOSH, 1999  Restricted distribution of loss-of-function mutations within the LMX1B genes of nail-patella syndrome patients. Hum. Mutat. 14:459-465[Medline].

DRAKE, J. W., B. CHARLESWORTH, D. CHARLESWORTH, and J. F. CROW, 1998  Rates of spontaneous mutation. Genetics 148:1667-1686[Abstract/Free Full Text].

DROST, J. B. and W. R. LEE, 1995  Biological basis of germline mutation: comparisons of spontaneous germline mutation rates among drosophila, mouse, and human. Environ. Mol. Mutagen. 25:48-64.

EYRE-WALKER, A. and P. D. KEIGHTLEY, 1999  High genomic deleterious mutation rates in hominids. Nature 397:344-347[Medline].

GIANNELLI, F., P. M. GREEN, S. S. SOMMER, M. C. POON, and M. LUDWIG et al., 1998  Haemophilia B: database of point mutations and short additions and deletions—eighth edition. Nucleic Acids Res. 26:265-268[Abstract/Free Full Text].

HARADA, K., S. KUSAKABE, T. YAMAZAKI, and T. MUKAI, 1993  Spontaneous mutation rates in null and band-morph mutations of enzyme loci in Drosophila melanogaster.. Jpn. J. Genet. 68:605-616[Medline].

KIM, N., J. KIM, D. PARK, C. ROSEN, D. DORSETT, and J. YIM, 1996  Structure and expression of wild-type and suppressible alleles of the Drosophila purple gene. Genetics 142:1157-1168[Abstract].

KONDRASHOV, A. S., 1988  Deleterious mutations and the evolution of sexual reproduction. Nature 336:435-440[Medline].

KONDRASHOV, A. S., 1998  Measuring spontaneous deleterious mutation process. Genetica 103:183-197.

KRANTZ, I. D., R. P. COLLITON, A. GENIN, E. B. RAND, and L. LI et al., 1998  Spectrum and frequency of Jagged1 (JAG1) mutations in Alagille syndrome patients and their families. Am. J. Hum. Genet. 62:1361-1369[Medline].

KRAWCZAK, M., E. V. BALL, I. FENTON, P. D. STENSON, and S. ABEYSINGHE et al., 2000  Human gene mutation database—a biomedical information and research resource. Hum. Mut. 15:45-51[Medline].

LEMAHIEU, V., J. M. GASTIER, and U. FRANCKE, 1999  Novel mutations in the Wiskott-Aldrich syndrome protein gene and their effects on transcriptional, translational, and clinical phenotypes. Hum. Mut. 14:54-66[Medline].

LLOYD, V. K., 1998 Drosophila melanogaster delta-adaptin subunit of AP-3 (garnet) gene, complete cds. GenBank/EMBL/DDBJ: AF044287.

LOHMANN, D. R., 1999  RB1 gene mutations in retinoblastoma. Hum. Mutat. 14:283-288[Medline].

LYKO, F., B. H. RAMSAHOYE, H. KASHEVSKY, M. TUDOR, and M. A. MASTRANGELO et al., 1999  Mammalian (cytosine-5) methyltransferases cause genomic DNA methylation and lethality in Drosophila.. Nat. Genet. 23:363-366[Medline].

MARTIN-MORRIS, L. E., K. LOUGHNEY, E. O. KERSHISNIK, G. POORTINGA, and S. HENIKOFF, 1993  Characterization of sequences responsible for trans-inactivation of the Drosophila brown gene. Cold Spring Harbor Symp. Quant. Biol. 58:577-584[Abstract/Free Full Text].

MAYER, K., W. BALLHAUSEN, and H.-D. ROTT, 1999  Mutation screening of the entire coding regions of the TSC1 and the TSC2 gene with the protein truncation test (PTT) identifies frequent splicing defects. Hum. Mutat. 14:401-411[Medline].

MOHRENWEISER, H., 1994  Impact of the molecular spectrum of mutational lesions on estimates of germinal gene-mutation rates. Mutat. Res. 304:119-137[Medline].

MUKAI, T. and C. C. COCKERHAM, 1977  Spontaneous mutation rates at enzyme loci in Drosophila melanogaster.. Proc. Natl. Acad. Sci. USA 74:2514-2517[Abstract/Free Full Text].

NACHMAN, M. W. and S. L. CROWELL, 2000  Estimate of the mutation rates per nucleotide in humans. Genetics 156:297-304[Abstract/Free Full Text].

NIIDA, Y., N. LAWRENCE-SMITH, A. BANWELL, E. HAMMER, and J. LEWIS et al., 1999  Analysis of both TSC1 and TSC2 for germline mutations in 126 unrelated patients with tuberous sclerosis. Hum. Mutat. 14:412-422[Medline].

NITASAKA, E., T. YAMAZAKI, and M. M. GREEN, 1995  The molecular analysis of brown eye color mutations isolated from geographically discrete populations of Drosophila melanogaster.. Mol. Gen. Genet. 247:164-168[Medline].

O'HARE, K., C. MURPHY, R. LEVIS, and G. M. RUBIN, 1984  DNA sequence of the white locus of Drosophila melanogaster.. J. Mol. Biol. 180:437-455[Medline].

PETROV, D. A. and D. L. HARTL, 1998  High rate of DNA loss in the Drosophila melanogaster and Drosophila virilis species groups. Mol. Biol. Evol. 15:293-302[Abstract].

PETROV, D. A. and D. L. HARTL, 1999  Patterns of nucleotide substitution in Drosophila and mammalian genomes. Proc. Natl. Acad. Sci. USA 96:1475-1479[Abstract/Free Full Text].

PETROV, D. A., E. R. LOZOVSKAYA, and D. L. HARTL, 1996  High intrinsic rate of DNA loss in Drosophila.. Nature 384:346-349[Medline].

PRITCHARD, J. K. and S. W. SCHAEFFER, 1997  Polymorphism and divergence at a Drosophila pseudogene locus. Genetics 147:199-208[Abstract].

PROSSER, J. and V. VAN HEYNINGEN, 1998  PAX6 mutations reviewed. Hum. Mutat. 11:93-108[Medline].

RUSSELL, L. B., 1999  Significance of the perigametic interval as a major source of spontaneous mutations that result in mosaics. Environ. Mol. Mutagen. 34:16-23[Medline].

RUSSELL, L. B. and W. L. RUSSELL, 1996  Spontaneous mutations recovered as mosaics in the mouse specific-locus test. Proc. Natl. Acad. Sci. USA 93:13072-13077[Abstract/Free Full Text].

SAMBROOK, J., E. F. FRITSCH and T. MANIATIS, 1989 Molecular Cloning: A Laboratory Manual, Ed. 2. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.

SELBY, P. B., 1998a  Major impacts of gonadal mosaicism on hereditary risk estimation, origin of hereditary diseases, and evolution. Genetica 102(103):445-462.

SELBY, P. B., 1998b  Discovery of numerous clusters of spontaneous mutations in the specific-locus test in mice necessitates major increases in estimates of doubling doses. Genetica 102(103):463-487.

SCHALET, A. P., 1960 A study of spontaneous visible mutations in Drosophila melanogaster. Ph.D. Thesis, Indiana University, Bloomington, IN.

SCHALET, A. P., 1986  The distribution of and complementation relationships between spontaneous x-linked recessive lethal mutations recovered from crossing long-term laboratory stocks of Drosophila melanogaster.. Mutat. Res. 163:115-144.

SEARLES, L. L., R. S. RUTH, A. M. PRET, R. A. FRIDELL, and A. J. ALI, 1990  Structure and transcription of the Drosophila melanogaster vermilion gene and several mutant alleles. Mol. Cell. Biol. 10:1423-1431[Abstract/Free Full Text].

SHABALINA, S. A. and A. S. KONDRASHOV, 1999  Pattern of selective constraint in C. elegans and C. briggsae genomes. Genet. Res. 74:23-30[Medline].

SHIANG, R., L. M. THOMPSON, Y. Z. ZHU, D. M. CHURCH, and T. J. FIELDER et al., 1994  Mutations in the transmembrane domain of FGFR3 cause the most common genetic form of dwarfism, Achondroplasia. Cell 78:335-342[Medline].

SOMMER, S. S., 1995  Recent human germ-line mutation: inferences from patients with hemophilia B. Trends Genet. 11:141-147[Medline].

TEARLE, R. G., J. M. BELOTE, M. MCKEOWN, B. S. BAKER, and A. J. HOWELLS, 1989  Cloning and characterization of the scarlet gene of Drosophila melanogaster.. Genetics 122:595-606[Abstract/Free Full Text].

TEN HAVE, J. F. M., M. M. GREEN, and A. J. HOWELLS, 1995  Molecular characterization of spontaneous mutations at the scarlet locus of Drosophila melanogaster.. Mol. Gen. Genet. 249:673-683[Medline].

TENG, D. H. F., L. B. BENDER, C. M. ENGELE, S. TSUBOTA, and T. VENKATESH, 1991  Isolation and characterization of the prune locus of Drosophila melanogaster.. Genetics 128:373-380[Abstract].

THOMPSON, J. N., R. C. WOODRUFF, and H. Y. HUAI, 1998  Mutation rate: a simple concept has become complex. Environ. Mol. Mutagen. 32:292-300[Medline].

TUFFERY-GIRAUD, S., S. CHAMBERT, J. DEMAILLE, and M. CLAUSTRES, 1999  Point mutations in the dystrophin gene: evidence for frequent use of cryptic splice sites as a result of splicing defects. Hum. Mutat. 14:359-368[Medline].

VOELKER, R. A., H. E. SCHAFFER, and T. MUKAI, 1980  Spontaneous allozyme mutations in Drosophila melanogaster: rate of occurrence and nature of the mutants. Genetics 94:961-968[Abstract/Free Full Text].

WARREN, W. D., S. PALMER, and A. J. HOWELLS, 1996  Molecular characterization of the cinnabar region of Drosophila melanogaster: identification of the cinnabar transcription unit. Genetica 98:249-262[Medline].

WESTERMAN, A. M., M. M. ENTIUS, P. P. C. BOOR, R. KOOLE, and E. DE BAAR et al., 1999  Novel mutations in the LKB1/STK11 gene in Dutch Peutz-Jeghers families. Hum. Mutat. 13:476-481[Medline].

WOODRUFF, R. C. and J. N. THOMPSON, JR., 1992  Have premeiotic clusters of mutation been overlooked in evolutionary theory? J. Evol. Biol. 5:457-464.

WOODRUFF, R. C., B. SLATKO and J. N. THOMPSON, JR., 1983 Factors affecting mutation rates in natural populations, pp. 37–124 in The Genetics and Biology of Drosophila, Vol. 3C, edited by M. ASHBURNER, H. L. CARSON and J. N. THOMPSON, JR. Academic Press, NY.

YAMAGUCHI, Y., T. S. TAKANO, T. YAMAZAKI, and K. HARADA, 1994  Molecular analysis of Gpdh null mutations that arose in mutation accumulation experiments in Drosophila melanogaster.. Heredity 73:397-404.