|Home | About | Journals | Submit | Contact Us | Français|
Coloration is one of the most variable characters in animals and provides rich material for studying the developmental genetic basis of pigment patterns. In the silkworm, more than 100 gene mutation systems are related to aberrant color patterns. The melanism (mln) is a rare body color mutant that exhibits an easily distinguishable phenotype in both larval and adult silkworms. By positional cloning, we identified the candidate gene of the mln locus, Bm-iAANAT, whose homologous gene (Dat) converts dopamine into N-acetyldopamine, a precursor for N-acetyldopamine sclerotin in Drosophila. In the mln mutant, two types of abnormal Bm-iAANAT transcripts were identified, whose expression levels are markedly lower than the wild type (WT). Moreover, dopamine content was approximately twice as high in the sclerified tissues (head, thoracic legs, and anal plate) of the mutant as in WT, resulting in phenotypic differences between the two. Quantitative reverse transcription PCR analyses showed that other genes involved in the melanin metabolism pathway were regulated by the aberrant Bm-iAANAT activity in mln mutant in different ways and degrees. We therefore propose that greater accumulation of dopamine results from the functional deficiency of Bm-iAANAT in the mutant, causing a darker pattern in the sclerified regions than in the WT. In summary, our results indicate that Bm-iAANAT is responsible for the color pattern of the silkworm mutant, mln. To our knowledge, this is the first report showing a role for arylalkylamine-N-acetyltransferases in color pattern mutation in Lepidoptera.
Coloration is one of the most variable traits in insects, and pigmentation is known to play a role in mimicry, sexual selection, thermoregulation, and other adaptive processes in many insect groups (1). More recently, there has been much interest among biologists in the molecular mechanisms underlying the great diversity of insect colors and color patterns.
Melanin, the dark pigment found in melanophores, is an important class of insect pigments (2,–4). The genes responsible for pigment metabolism have been systematically studied in many insects, including Drosophila melanogaster, Manduca sexta, and Papilio xuthus (2, 3, 5,–8). The melanin is derived from tyrosine (10, 11). Tyrosine hydroxylase and dopa decarboxylase convert tyrosine into dopa and dopa into dopamine, respectively (2). Dopa serves as the precursor of dopa melanin (which is black) and is converted to dopa melanin by the activities of the Yellow, Yellow-f 1 and Yellow-f 2 proteins (12). Furthermore, dopamine serves as the precursor of dopamine melanin (black or brown in color and a major pigment of the insect cuticle), the change being caused by the catalytic action of phenol oxidases (1, 2, 3, 5, 6, 12,–14). The current study confirmed that the dopamine content was elevated in melanized insects or in the melanized regions of insects (5, 15). Alternatively, dopamine can reversibly convert to N-β-alanyldopamine (NBAD,3 yellowish) by NBAD synthetase (EBONY) and NBAD hydrolase. NBAD deposited in the cuticle of an insect causes a yellow coloration. Finally, a family of arylalkylamine-N-acetyltransferases (AANATs) convert dopamine into N-acetyldopamine, a precursor of the colorless N-acetyldopamine sclerotin (2, 3). In insects, AANATs are essential for sclerotization of the cuticle and catabolism of monoamines (16, 17). However, they might also be related to melanism in insects (15). In addition, the activities of AANAT and melatonin display day/night fluctuations in the head tissues of the silkworm (Bombyx mori) and in cockroach (Periplaneta americana) (18,–20). Tsugehara et al. (21) cloned and characterized full-length cDNA of Bm-iAANAT in silkworm. At present, nothing is known about the relationship between Bm-iAANAT and pigmentation metabolism or pigment mutants.
Silkworm, B. mori, is an important economic insect and lepidopteran molecular model (22). More than 600 mutant systems are maintained in the silkworm gene bank at Southwest University (Chongqing, China). More than 100 of these are related to body color and body markings (23), and they are distributed over all of the silkworm linkage groups except the 27th and 28th. Therefore, silkworm provides an excellent model for studying the genetic and molecular mechanisms that generate varied pigment patterns. Recently, the candidate genes of some color pattern mutants such as ch, so (7), lem (24), and rb (25) have been cloned. However, the molecular bases of many other mutations are still unknown.
The mln (melanism) is a rare body color mutant that exhibits a clearly recognizable phenotype both in larval and adult B. mori. It is regulated by a single recessive gene and has been mapped at 41.5 cM on the 18th linkage in the silkworm genetic linkage map. In the recessive homozygote of the mln mutant, the larval head, anal plate, thoracic legs, and sieve plates of the spiracles are black. In the adult, the body, wings, and integument are also black instead of the normal color pattern (Fig. 1). Excess melanin appears to accumulate in the mln mutant. However, the candidate gene responsible has not been identified previously.
In the present study, we fine mapped the mln locus and identified the candidate gene responsible, Bm-iAANAT. In addition, we found differences between Dazao and Dazao-mln in their gene sequences and transcription levels of Bm-iAANTA and investigated the differences in their expression patterns of other key genes involved in the melanin metabolism pathway. Based on our experimental results, we conclude that the Bm-iAANAT gene plays an essential role in the pigment metabolism in silkworm and that the abnormal Bm-iAANAT is responsible for the mln mutant.
The melanism mutant strain Dazao-mln (near isogenic line of mln), wild-type (WT) strain Dazao and C108 were obtained from the silkworm gene bank in Southwest University. The silkworms were reared on mulberry leaves under a photoperiod of 12 h/12 h light/dark at 25 °C during the experiments.
Two silkworm strains, C108 (+mln/+mln) and Dazao-mln (mln/mln) were used for preliminary mapping. A single-pair cross between a female (C108) and a male (Daza-mln) produced the F1 offspring. For linkage analysis, 20 BC1F progeny (10 wild-type and 10 melanism offspring) from the cross (C108 × Dazao-mln) ♀ × Dazao-mln were used, although another backcross, Dazao-mln♀ × (C108 × Dazao-mln) was used for recombination analysis. SSR (simple sequence repeat) markers were obtained from the published SSRs linkage map (26). Segregation patterns were analyzed using Mapmaker/exp (27) and the Kosambi mapping function (28).
Based on the results of the preliminary SSR mapping, we analyzed the upstream and downstream sequences close to the tightly linked SSR marker of the mln locus. This program requires the silkworm 9× assembly genome database (29), BLAST (30), and primer 5.0. We used the predicted gene fragments as query sequences to search the NCBI database with the BLASTp program and found the homolog and ortholog genes that have been shown to be involved in the pigmentation pathway in other species. We then designed primers as new markers based on the genome sequences of the predicted candidate gene. The genome fragment of candidate gene that exhibited polymorphism between the parents was used for fine mapping. The primers are shown in supplemental Tables 1 and 2.
The total RNAs were extracted from Dazao and Dazao-mln on day 4 of the 5th instar. The primers (supplemental Table 3) for the full-length cDNA of the candidate gene were designed according to the sequence in GenBank (21). The PCR products were cloned into the pMD19-T vector (Takara) and sequenced.
To investigate the expression patterns, we used semiquantitative RT-PCR to analyze total RNAs isolated from the whole body of several developmental stages (from day 3 of the 4th instar larvae to eclosion of the adult). Total RNAs were purified using TRIzol (Invitrogen) according to the manufacturer's protocol and subjected to cDNA synthesis using oligo(dT) primers and a Moloney murine leukemia virus reverse transcriptase (Promega) according to the manufacturer's instructions. Ten silkworm tissues were sampled on day 4 of the 5th instar larvae of Dazao, including the silk gland, midgut, malpighian tubule, fat body, testis, ovary, head, integument, hemocyte, and trachea and used to investigate the pattern of spatial expression. The primers designed for RT-PCR are shown in supplemental Tables 4. The BmActin3 gene was used as an internal control. The expression levels of the candidate gene in sclerified (anal plate, thoracic legs, and head) and less sclerified tissues (epidermis) of the Dazao larvae were detected using quantitative RT-PCR.
We performed the quantitative RT-PCR using the ABI Prism 7000 sequence detection system (Applied Biosystems) with a SYBR Premix EX Taq kit (Takara) to detect the expression levels of the Bm-iAANAT gene in both Dazao (WT) and Dazao-mln (mutant) at the same developmental stages and in the same sampled tissues (head, thoracic legs, and anal plate). The quantitative RT-PCR primers used are shown in supplemental Tables 5. Eukaryotic translation initiation factor 4A (silkworm microarray probe ID: sw22934) was used as an internal control.
Referring to the melanin metabolism pathway in D. melanogaster (2, 3) and the full-length cDNA sequences of Pale (AB439286), Yellow (AB438999), Ddc (AF372836), and ebony (AB439000) in silkworm, we designed primers for use in quantitative RT-PCR (supplemental Table 6). The sw22934 was used as an internal control.
Dopamine was extracted from the strain-specific phenotype tissues (head, thoracic legs, and anal plate: total quantity, 200 mg.) on day 2 of 4th instar mln and WT. The tissues were homogenized in 0.5 ml of 1.2 m HCl containing 5 mm ascorbic acid in a centrifuge tube and centrifuged at 14,000 rpm, 4 °C for 10 min. Supernatants were incubated at 100 °C for 10 min and then added to 0.5 ml chloroform and shaken before further centrifugation at 12,000 rpm at 4 °C for 10 min. Supernatants were analyzed by HPLC. Chromatography was performed on a Waters 1525 binary HPLC equipped with a Waters 2487 dual λ absorbance detector and a SymmetryShield RP18 (5 μm, 4.6 × 150-mm) column. The flow velocity of the mobile phases (69.5 mm KH2PO4, 5.5 mm K2HPO4 (pH 5.8):methanol (Sigma; chromatography pure) 98:2) was 1 ml/min, isocratic elution. The column elution was monitored at 280 nm. The standard sample was dopamine hydrochloride (Sigma, supplemental Fig. 2).
To determine the chromosomal location of the mln locus, we performed genetic linkage analysis. First, using the microsatellite (SSR) markers on the 18th linkage in the silkworm SSR molecular linkage map, we roughly mapped the mln mutation using 372 BC1M individuals. The preliminary mapping showed that the marker S1807 was tightly linked with the mln locus (Fig. 2). We analyzed the upstream and downstream sequences (~300 kb on each side) of S1807 in the silkworm database (31). Within ~100 kb upstream of the S1807 marker, we found two genes (BGIBMGA008538 and BGIBMGA008539) encoding AANAT which plays an important role in the synthesis of N-acetyldopamine in the pigmentation pathway of D. melanogaster. We designed primer sets based on the sequences of the two genes to detect the genomes of Dazao-mln and C108. One pair of primers, named P3C, from BGIBMGA008538, displayed a smaller amplification product in Dazao-mln than in C108 and showed the same polymorphism between Dazao-mln and its recurrent parent, Dazao. We used P3C as a new marker for fine mapping. The result suggested that there was no recombination between P3C and the mln locus (Fig. 2). Furthermore, the sequences amplified by the primers of P3C are partly located on the exon of the BGIBMGA008538, which has been previously identified as Bm-iAANAT (GenBank accession number NM_001079654.1). Therefore, we concluded that the Bm-iAANAT gene was the candidate gene for the melanism (mln) mutant.
Cloning and comparing the sequences of the P3C-amplified region in Dazao, C108, and Dazao-mln, we found that it was 410 bp and 506 bp in the Dazao-mln and wild-type strains (C108, Dazao), respectively. The mutation deleted 96 bp and inserted another 29-bp fragment compared with the WT (Fig. 3A). In addition, we cloned the complete Bm-iAANAT cDNA sequences of the WT and mln mutant. A significant amount of a 1460-bp fragment was amplified from the WT (Fig. 3B). Compared with the genome assembly data, it was apparent that Bm-iAANAT has 5 exons encoding 261 amino acids. However, two types of abnormal transcript, type 1 cDNA (1253 bp) and type 2 cDNA (1407 bp), were amplified in the Dazao-mln mutant (Fig. 3B). As shown schematically in Fig. 3C, the type 1 cDNA lacked almost the whole 4th exon except for the first 2 bp and causes a premature TAG stop codon not found in the WT. Type 2 lacked the last 67 bp of the 4th exon, and recruited a 15-bp length intron sequence. The termination codon of type 2 was located on the 3′-untranslated region of the normal transcript (Fig. 3C). Thus, type 1 and type 2 both encoded for potentially aberrant proteins and destroyed the acetyltransferase 1 domain compared with the WT (Fig. 3D).
Temporal expression patterns of the Bm-iAANAT gene were detected in the whole body of Dazao during the developmental period from the 3rd day of the 4th instar larvae to eclosion of the adult (Fig. 4, A and B). The semi-RT-PCR analysis showed that the expression levels of Bm-iAANAT fluctuated during the development period. It was expressed normally during the 3rd day of the 4th instar larvae (intermolt), turned off at the beginning of the fourth molting stage (from 0 to 16 h), and then reappeared in the postmolt stage (24 h later) (Fig. 4A). Relatively high expression levels were maintained during most of the 5th instar larval period, and no expression was detected at the commencement of the wandering stage. A sharp increase followed at the end of the wandering stage (W2.5 day), which is opposite to the change of a 20E titer at the same time (32). The greatest expression appeared on the seventh (P7) and eighth (P8) day following pupation (Fig. 4B).
To investigate the spatial expression patterns of the Bm-iAANAT gene, we also used semi-RT-PCR to analyze total RNA in samples of 10 silkworm tissues from 4th day 5th instar larvae of Dazao. The results showed that the Bm-iAANAT gene is expressed in the head, silk gland, and integument. The highest level of expression was found in the head. In other tissues, its expression was very low or nearly undetectable (Fig. 4C).
To examine the relationship between the expression of Bm-iAANAT and the body regions where the mln strain exhibited a different color compared with the WT, we performed a quantitative RT-PCR analysis using cDNA prepared from the less sclerified epidermis and the more sclerified head, thoracic legs, and anal plate of 5th instar Dazao larvae. The Bm-iAANAT was found to be expressed at higher levels in sclerified tissues (head, thoracic legs, and anal plate) than in less sclerified tissue (epidermis) (Fig. 4D). In other words, high expression of Bm-iAANAT correlated with the black-colored regions but not with the non—black-colored region in the mln mutant.
We compared the expression of Bm-iAANAT between Dazao-mln and Dazao at several developmental stages, including the 4th day of the 5th instar larva (V4), the 2nd day of pupation (P2), and at the eclosion of the adult (M0), when Bm-iAANAT normally showed its highest activity. The amplified sequences were shared by the three types of Bm-iAANAT cDNA. The result of the semiquantitative RT-PCR showed that the level of expression of the two transcripts in the mln mutant was lower than that of the normal transcript in Dazao (Fig. 5A). As a more rigorous test, we performed a quantitative RT-PCR analysis with the same materials and showed that the level of expression of normal Bm-iAANAT in WT is about 10–20 times higher than that of the two transcripts in the mln mutant (Fig. 5B). In addition, the expression level of Bm-iAANAT in Dazao was also higher than that in the mln mutant in the regions which exhibit obvious phenotypic differences between WT and mln (head, thoracic legs, and anal plate) (Fig. 5C).
To test how other genes involved in the melanin metabolism pathway respond to Bm-iAANAT level changes, we performed a quantitative RT-PCR analysis to detect the expression levels of several genes (Pale, DdC, Yellow, and ebony) in the regions with different Dazao and Dazao-mln coloration (head, thoracic legs, and anal plate). The results show that expression of Pale, DdC, and Yellow in these regions were all higher in Dazao than in Dazao-mln, whereas the expression of ebony was lower in Dazao than in Dazao-mln, except in the head (Fig. 6A). In the adult moth, the expression of all four of the above genes was higher in Dazao than in Dazao-mln, the ratios of relative expression levels (Dazao-mln/Dazao) of DdC and ebony being 0.471 and 0.573, respectively (Fig. 6B).
The result of a reversed phase HPLC revealed that the content of dopamine in the mln mutant (143.233 ± 11.821 μg/g) was ~2 times higher than that in the WT (72.953 ± 2.022 μg/g) (Fig. 7). Therefore, we concluded that a greater accumulation of dopamine resulted from the functional deficiency of Bm-iAANAT in the mutant and that the excessive dopamine was converted into dopamine melanin, causing the darker color pattern of the sclerified regions in the mln mutant compared with the WT.
In this study, we fine map the candidate gene of the mln mutant using SSR markers and material from the silkworm genome database. We confirmed the results using the Dazao and the near isogenic line Dazao-mln obtained by backcrossing to a Dazao strain for more than 20 generations. This provided a powerful combination of genetic material with which to map the candidate gene. In the fine mapping, we focused on the Bm-iAANAT gene. By cloning the transcripts of the WT (Dazao) and mln mutant (Dazao-mln), we found that two abnormal transcripts were present in the Dazao-mln compared with the WT (Dazao). The exact sequence change of Bm-iAANAT genes in the mutant strongly suggested that it would be responsible for the mln mutation.
At the beginning of 4th molting stage, the titer of ecdysteroid (20-hydroxyecdysone, 20E), the major active molting hormone, increased rapidly and reached a peak at about the 14th h and dropped sharply 20 h later, and then the 4th larval molting process was completed (33, 34). Interestingly, the expression of Bm-iAANAT seemed to be suppressed by the 20E in the same manner as has been found in Drosophila (35). The situation was consistent with those of other melanin synthesis enzymes (tyrosine hydroxylase, dopa decarboxylase, Yellow, EBONY), which are known to be regulated by ecdysteroid titer during the molting process (4). The expression of Bm-iAANAT reached a peak at the 7th (P7) and 8th (P8) day after pupation, when the moth cuticle and scales covering the wing and body were formed. Generally, the massively high expression levels of the Bm-iAANAT gene occurred once the new epidermis had formed and pigmentation and exocuticle secretion began. Furthermore, the spatial expression analysis revealed that Bm-iAANAT was expressed heavily in the sclerified tissues (head, thoracic legs, and anal plate) where the black coloration occurs in the mln strain. We therefore concluded that the abundant Bm-iAANAT expression promotes accumulation of the NADA necessary for cuticle sclerotization and that the consumption of dopamine follows the same course as in Drosophila.
In the mln mutant, the two types of abnormal Bm-iAANAT transcript could not code the functional enzyme, and the expression levels of Bm-iAANAT were much lower than that in Dazao (WT) (Fig. 5). Consequently, the excess dopamine could not be consumed in the sclerified tissues, producing the black pigment (dopamine melanin). This presumption was confirmed by measuring the dopamine levels, which were two times higher in the mln mutant than that in WT (Fig. 7). Therefore, the corresponding sclerified parts of the larva (head, thoracic legs, anal plate, sieve plates of the spiracle, and claw hook of prolegs), the body color (tentacles, veins, dorsal plate, and thoracic legs), and the scales of the adult in the mln mutant are blacker or more buffy colored than in Dazao (WT). It is therefore intriguing that the abnormal pigmentation appears only in certain parts of the body (the sclerified regions) rather than more generally. Furthermore, other genes involved in the melanin metabolic pathway were also regulated by the loss of function of Bm-iAANAT in the mln mutant (Fig. 6C). The expression of pale, Yellow, and, especially, Ddc were all suppressed by the accumulation of dopamine in three particular tissues (head, thoracic legs, and anal plate) in Dazao-mln larva. However, ebony was up-regulated in the thoracic legs and anal plate, resulting from the excess dopamine, and extra NBAD generated. This may explain the buffy colors occurring in the thoracic legs and anal plate of the mutant (Figs. 1A, ,6,6, A and C, and supplemental Fig. 1). However, we do not know why the ebony was down-regulated in the head. In the adult stage, all of the four key genes were down-regulated in the mln mutant, and we speculate that Ddc, pale, and Yellow were also suppressed by the accumulation of dopamine (Figs. 1B, ,6,6, B and C, and supplemental Fig. 1). However, the lower expression of ebony may be due to less NBAD being needed to produce a yellow color in the mln mutant (9). Further studies are required to test this hypothesis.
In summary, several lines of evidence suggest that Bm-iAANAT is responsible for the mutant silkworm color pattern, mln. To our knowledge, this is the first report showing a role for Bm-iAANAT in the color pattern besides its activity in synthesizing melatonin in the silkworm (18). In general, the body color patterns of insects are determined by complicated genetic pathways, and recently many studies have looked in detail at this phenomenon. There is no doubt that elucidating the molecular basis of these color pattern mutants will be useful in illuminating the development and evolution of insect coloration.
*This work was supported by National Basic Research 973 Program of China Grant 2005CB121000, Hi-Tech Research and Development 863 Program of China Grant 2006AA10A117, National Natural Science Foundation of China Grant 30671591, Science Foundation of Ministry of Education of China Grant 20060635016, and Technological Innovative Foundation Project for postgraduates of Southwest University of China Grant 2006052.
The nucleotide sequences reported in this paper have been submitted to the Gen-BankTM/EBI Data Bank with accession number(s) GQ169236 and GQ169237.
3The abbreviations used are: