Hideaki Maekawa

Carotenoid silk coloration is controlled by a carotenoid-binding protein, a product of the Yellow blood gene.

Mechanisms for the uptake and transport of carotenoids, essential nutrients for humans, are not well understood in any animal system. The Y (Yellow blood) gene, a critical cocoon color determinant in the silkworm Bombyx mori, controls the uptake of carotenoids into the intestinal mucosa and the silk gland. Here we provide evidence that the Y gene corresponds to the intracellular carotenoid-binding protein (CBP) gene. In the Y recessive strain, the absence of an exon, likely due to an incorrect mRNA splicing caused by a transposon-associated genomic deletion, generates a nonfunctional CBP mRNA, resulting in colorless hemolymph and white cocoons. Enhancement of carotenoid uptake and coloration of the white cocoon was achieved by germ-line transformation with the CBP gene. This study demonstrates the existence of a genetically facilitated intracellular process beyond passive diffusion for carotenoid uptake in the animal phyla, and paves the way for modulating silk color and lipid content through genetic engineering.

A carotenoid-binding protein (CBP) plays a crucial role in cocoon pigmentation of silkworm (Bombyx mori) larvae

We examined the role of carotenoid‐binding protein (CBP) in yellow cocoon pigmentation. First, using yellow or white cocoon races, we investigated the linkage between the yellow pigmentation and CBP expression. CBP was expressed only in the silk gland of the yellow cocoon races, which utilize carotenoids for cocoon pigmentation. Furthermore, CBP expression in the silk glands of day 1–7 fifth instar larvae matched the period of carotenoid uptake into the silk gland. Finally, we gave double‐stranded CBP RNA to Bombyx mori (B. mori) larvae to induce RNA interference. The significantly reduced expression of CBP in the silk gland of fifth instar larva was confirmed on day 4 and a decrease in yellow pigmentation was observed in the cocoon. We showed that CBP plays a key role in the yellow cocoon pigmentation caused by carotenoids.

Isolation, Characterization, and cDNA Sequence of a Carotenoid Binding Protein from the Silk Gland of Bombyx mori Larvae

A carotenoid binding protein (CBP) has been isolated from the silk glands of Bombyx mori larvae. The protein has an apparent molecular mass of 33 kDa and binds carotenoids in a 1:1 molar ratio. Lutein accounts for 90% of the bound carotenoids, whereas α-carotene and β-carotene are minor components. Immunological analysis demonstrated the presence of CBP only in the yellow-colored tissues of the silk gland, midgut, testis, and ovary. Several phenotypes of B. mori mutants linked to carotenoid transport have been utilized to characterize CBP. TheY (yellow hemolymph) gene controls uptake of carotenoids from the midgut lumen into the midgut epithelium, and larvae with the + Y gene lack this property. Immunoblotting analysis confirmed the presence of CBP in mutants with the dominant Y gene only. Immunohistochemistry verified the localization of CBP in the villi of the midgut epithelium, indicating that CBP might be involved in absorption of carotenoids. A cDNA clone for CBP encoding a protein of 297 amino acids has been isolated from the B. mori silk gland cDNA library. The deduced amino acid sequence revealed that CBP is a novel member of the steroidogenic acute regulatory (StAR) protein family with its unique structural feature of a StAR-related lipid transfer domain, known to aid in lipid transfer and recognition. Lutein-binding capacity of the recombinant CBP (rCBP) determined by incubating rCBP with lutein followed by immunoprecipitation using anti-CBP IgG conjugated to protein A-Sepharose, demonstrated the formation of a lutein-rCBP complex. Sequence analyses coupled with binding specificity suggest that CBP is a new member of the StAR protein family that binds carotenoids rather than cholesterol.

Lipid transfer particle catalyzes transfer of carotenoids between lipophorins of Bombyx mori.

The yellow color of Bombyx mori hemolymph is due to the presence of carotenoids, which are primarily associated with lipophorin particles. Carotenoids were extracted from high density lipophorin (HDLp) of B. mori and analyzed by HPLC. HDLp contained 33 micrograms of carotenoids per mg protein. Over 90% of carotenoids were lutein while alpha-carotene and beta-carotene were minor components. When larval hemolymph was subjected to density gradient ultracentrifugation, a second minor yellow band was present, which was identified as B. mori lipid transfer particle (LTP). During other life stages examined however, this second band was not visible. To determine if coloration of LTP may fluctuate during development, we determined its concentration in hemolymph and compared it to that of lipophorin. Both proteins were present during all life stages and their concentrations gradually increased. The ratio of lipophorin: LTP was 10-15:1 during the fourth and fifth instar larval stages, and 20-30:1 during the pupal and adult stages. Thus, there was no correlation between the yellow color attributed to LTP and its hemolymph concentration. It is possible that yellow coloration of the LTP fraction corresponds to developmental stages when the particle is active in carotene transport. To determine if LTP is capable of facilitating carotene transfer, we took advantage of a white hemolymph B. mori strain which, when fed artificial diet containing a low carotene content, gives rise to a lipophorin that is nearly colorless. A spectrophotometric, carotene specific, transfer assay was developed which employed wild type, carotene-rich HDLp as donor particle and colorless low density lipophorin, derived from the white hemolymph strain animals, as acceptor particle. In incubations lacking LTP carotenes remained associated with HDLp while inclusion of LTP induced a redistribution of carotenes between the donor and acceptor in a time and concentration dependent manner. Time course studies suggested the rate of LTP-mediated carotene transfer was relatively slow, requiring up to 4 h to reach equilibrium. By contrast, studies employing 3H-diacylglycerol labeled HDLp as donor particle in lipid transfer assays revealed a rapid equilibration of label between the particles. Thus, it is plausible that the slower rate of LTP-mediated carotene transfer is due to its probable sequestration in the core of HDLp.

A defective non-LTR retrotransposon is dispersed throughout the genome of the silkworm, Bombyx mori

The presence of long repetitive sequences is demonstrated in the genome of the silkworm, Bombyx mori. Members of this BMC1 family reveal several features typical of the L1 (long interspersed sequence one) family of mammals, except for species specific elements. The number of BMC1 elements is estimated to be approximately 3500 per haploid genome. Elements containing the full length unit of 5.1 kb are dispersed throughout the genome and their restriction sites are conserved, although most members are preferentially truncated to varying extents at their 5′ ends. DNA sequencing indicates that this element contains six tandem repeats of 15 bp CpG-rich sequence in the 5′ proximal region. It terminates with a 3′ oligo(A) stretch, and is flanked at both ends by a 7–10 bp target sequence duplication. In addition, there is significant evidence for amino acid sequence homology with reverse transcriptase domains of other L1 families, especially F, Doc and lockey of Drosophila melanogaster. No large open reading frame is present. The BMC1 element is suggested to be dispersed in the genome by a transposition mechanism involving RNA intermediates.

Transfer of cholesterol and diacylglycerol from lipophorin to Bombyx mori ovarioles in vitro: role of the lipid transfer particle

The objective of this study was to characterize the transfer of diacylglycerol (DAG) and cholesterol from larval Bombyx mori lipophorin to ovarioles. Transfer studies were carried out by incubating pupal ovarioles (5-day) with [(3)H]-cholesterol and [(3)H]-DAG-labeled lipophorin under different conditions. Transfer of both cholesterol and DAG exhibited hyperbolic dependency on lipophorin concentration with apparent Km values of 0.83 +/- 0.17 mg/ml and 0.74 +/- 0.16 mg/ml, respectively. Pretreatment of ovarioles with anti-lipid transfer particle (LTP) IgG significantly inhibited transfer of labeled DAG to ovarioles (75%) and not cholesterol. Injection of B. mori pupae (day 4) with anti-LTP IgG significantly affected the weight (65%), number of eggs (49%), amount of lipid (74%), and protein (65%) of the adult ovaries. Matured eggs had a very faint yellow color and deformed shape compared to controls. The inhibitory effect demonstrates the active role LTP plays in growth of ovaries, development, and oogenesis. The effect on vitellogenin shortage on egg development and maturation was determined by implanting ovaries in male recipients that lack vitellogenin. An 80% decline in egg production was observed. However, the mature eggs were normal in shape, color, and lipid content. Thus, restricting lipid or protein delivery to developing ovaries would dramatically affect choriogenesis.

Purification and properties of a lipid transfer particle from Bombyx mori: comparison to the lipid transfer particle from Manduca sexta

A lipid transfer particle (LTP) was purified from the hemolymph of the silkworm Bombyx mori. Like other insect LTPs, the B. mori LTP is a very high density lipoprotein containing 21% lipid and three apoproteins of mass approximately 350 kDa, approximately 85 kDa, and approximately 60 kDa. B. mori LTP catalyzes the exchange of lipids between different density class lipoproteins found in adult hemolymph and between adult lipoproteins and vitellogenin. However, in no case was net lipid transfer observed. Manduca sexta LTP also catalyzed exchange of lipids, but not net transfer of lipids, between different density class lipoproteins found in adult hemolymph.

Nucleolus organizers in the wild silkworm Bombyx mandarina and the domesticated silkworm B. mori

Two types (Ra1 and Ra2) of nucleolus organizers were identified in the genome of Bombyx mandarina (Japan) which occurs in Japan. Genetical analysis of a hybrid with B. mori suggested that the loci of both nucleolus organizers are allelic and correspond to the R0 locus of B. mori. These nucleolus organizers segregated and were inherited by the progeny in a Mendelian fashion. The majority of the Ra1 rDNA units were 10.6 kb in length and had an additional EcoRI site in the transcribed spacer region when compared with the same size unit of R0. On the other hand, the KpnI site present in the non-transcribed spacer region of the R0 rDNA was not detected in the Ra1 unit. The 15.1 kb unit observed in the Ra2 locus was the same as the unit with the type II intron of R0. The four major components of Ra2 rDNA, with lengths of 10.6, 15.1, 15.7 and 20.8 kb, were also found in the R0 locus, and thus the Ra2 and the R0 loci were considered to be closely related. Usually the functional unit of rDNA in the nucleolus organizers of homologous or non-homologous chromosomes cannot be easily distinguished. However, in the case of B. mandarina (Japan), distinct functional 10.6 kb units were observed in the allelic Ra1 and Ra2 loci. Therefore the existence of the two distinct units suggest the possibility of introducing the chromosomes of the interspecies in the genus Bombyx.

Purification and Molecular Cloning of Xylanases from the Wood-Feeding Termite, Coptotermes formosanus Shiraki

Coptotermes formosanus is one of the most destructive termites in the southern part of Japan as well as in the United States. Hemicellulose is a noncellulosic polysaccharide found in plant cell walls, and xylan is the major constituent of hemicellulose. Since hemicellulose prevents access of cellulolytic enzymes to cellulose, enzymatic hydrolysis of hemicellulose is beneficial for cellulose digestion. We purified three functional xylanases to homogeneity from C. formosanus for the first time. Elution profiles from the whole termite extract suggest that these three xylanases play major roles in xylan digestion in the gut of the termites. The corresponding cDNAs were successfully cloned based on the N-terminal amino acid sequences, encoding GHF11 xylanases. Reverse transcription-PCR using manipulated protozoan cells in the hindgut revealed that the corresponding genes were expressed in the symbiotic flagellate Holomastigotoides mirabile. These results suggest that the GHF11 xylanases that are produced by the symbiotic flagellates play a primary role in xylan degradation in C. formosanus.

Molecular dynamics simulation of clustered DNA damage sites containing 8-oxoguanine and abasic site.

Clustered DNA damage sites induced by ionizing radiation have been suggested to have serious consequences to organisms, such as cancer, due to their reduced probability to be repaired by the enzymatic repair machinery of the cell. Although experimental results have revealed that clustered DNA damage sites effectively retard the efficient function of repair enzymes, it remains unclear as to what particular factors influence this retardation. In this study, approaches based on molecular dynamics (MD) simulation have been applied to examine conformational changes and energetic properties of DNA molecules containing clustered damage sites consisting of two lesioned sites, namely 7,8‐dihydro‐8‐oxoguanine (8‐oxoG) and apurinic/apyrimidinic (AP) site, located within a few base pairs of each other. After 1 ns of MD simulation, one of the six DNA molecules containing a clustered damage site develops specific characteristic features: sharp bending at the lesioned site and weakening or complete loss of electrostatic interaction energy between 8‐oxoG and bases located on the complementary strand. From these results it is suggested that these changes would make it difficult for the repair enzyme to bind to the lesions within the clustered damage site and thereby result in a reduction of its repair capacity.