A high-frequency null mutant of an odorant-binding protein gene, Obp57e, in Drosophila melanogaster

Abstract

We have found a null mutant of an odorant-binding protein, Obp57e, in Drosophila melanogaster. This frameshift mutation, which is a 10-bp deletion in the coding region, is at a high frequency in the Kyoto population and is also present in Taiwan and Africa. We have sequenced a 1.5-kb region including the tandemly duplicated gene, Obp57d, from 16 inbred lines sampled in Kyoto, Japan. The analyses showed a peak of nucleotide diversity and strong linkage disequilibrium around this mutation. This pattern suggests an elevated mutation rate or an influence of balancing selection in this region. The level of nucleotide divergence between D. melanogaster and D. simulans does not support the former possibility. Thus, this presence/absence polymorphism may be due to balancing selection, which takes advantage of the relatively weak functional constraint in members of a large gene family. In addition, the Obp57d gene region showed an excess of high-frequency-derived mutants that is consistent with a pattern predicted under positive natural selection.

Figure 1.—

Physical map of the Obp57d and Obp57e regions (shown by arrows). Open rectangles, coding regions of the two genes. Open triangles, insertion/deletion. Numbers indicate the positions of the nucleotide variants detected in D. melanogaster from Kyoto, Japan. Numbers in rectangles and ovals indicate positions of amino acid changes and frameshift mutations, respectively.

Figure 2.—

Nucleotide variations detected in the Obp57d and Obp57e regions of 16 D. melanogaster lines from Kyoto, Japan. The first row indicates the nucleotide information of the sequenced D. simulans line. Nucleotide information of 7 other D. melanogaster lines is in the last rows. Numbers in rectangles and ovals indicate positions of amino acid changes and frameshift mutations, respectively. The presence and absence of the sequence at each indel position is indicated, respectively, by + and −.

Figure 3.—

RT-PCR products from (A) Obp57d and (B) Obp57e transcripts. Lanes indicate cDNA templates derived from the following: W, wings; T, tarsi; L, labela; A, antennae; M, maxillary palps. RNA was extracted from tissues of 10 3-day-old adults of both sexes. (C) Products of ribosomal protein 49 (Rp49) gene served as the internal control. All the primer pairs were designed to span introns.

Figure 4.—

Sliding-window analyses in the Obp57d and Obp57e regions. (A) Sliding-window profile of π. (B) Sliding-window profile of nucleotide diversity within (dotted lines) and between (solid line) wild-type and frameshift mutant classes. (C) Sliding-window profile of nucleotide divergence (average number of nucleotide substitutions per site) between D. melanogaster (16 lines from Kyoto) and D. simulans. Window size, 100 bp; step size 25 bp. Alignment gaps are excluded from the analyses.

Figure 5.—

(A) Linkage disequilibrium between polymorphisms in the Obp57d region (left) and the Obp57e region (right). All the phylogenetically informative sites (nonsingletons) are used for the analyses. Pairs of sites that are P < 0.05 by Fisher's exact test after Bonferroni correction are solid. (B) Linkage disequilibrium (R²) plotted against distance (in base pairs) between polymorphic sites in the Obp57d region (left) and the Obp57e region (right). Open circles represent pairs of sites that are P < 0.05 by Fisher's exact test after Bonferroni correction. The fitted lines are for the values of k that minimize the sums of squared deviations between the observed value of R² and its expectation: R² = 1/(1 + k × distance) + 1/16 (Weir and Hill 1986). k is a constant in the unit of 4N_ec. The best fit was k = 0.0049 for the Obp57d region and k = 0.0070 for the Obp57e region.

Figure 6.—

Frequency spectrum of the Obp57d region (left) and the Obp57e region (right). Open and solid bars indicate amino acid replacement sites and silent sites, respectively.

Figure 7.—

Sliding-window profile of nucleotide diversity within (dotted lines) and between (solid line) wild-type and frameshift mutant classes in African lines. The wild-type class consists of MEL6G59, MEL8, and KSA2 lines; the frameshift mutant class consists of MEL7, Algeria, and CA1 lines. Window size, 100 bp; step size, 25 bp, starting on nucleotide position 708. Alignment gaps are excluded from the analyses.