Heredity 82 (1999) 661±667
Received 7 October 1998, accepted 26 January 1999
The in¯uence of triploidy on gene expression in the silkworm, Bombyx mori MASATAKA G. SUZUKI à, TORU SHIMADA* , TAKESHI YOKOYAMA§ & MASAHIKO KOBAYASHI Department of Agricultural and Environmental Biology, Graduate School of Agricultural and Life Sciences, University of Tokyo, Yayoi, Bunkyo-ku, Tokyo 113±8657, Japan and §Department of Biological Production, Faculty of Agriculture, Tokyo University of Agriculture and Technology, Saiwai-cho, Fuchu, Tokyo 183, Japan
1 2 1
In Bombyx mori, it is well established that polyploids are easily induced when newly laid eggs are exposed to a variety of conditions, such as high or low temperature, centrifugal force, or chemicals like colchicine. To investigate gene dosage eects by varying the ploidy, the transcription levels of six genes expressed in various tissues were analysed in the diploid and two dierent genetically produced triploids (PPC and CCP). In the PPC triploid, the transcription level per cell of two genes was directly proportional to the structural gene dosage, whereas two other genes showed the mRNA level expected if compensation occurred. In the CCP triploid, three genes displayed dose-dependent levels of expression, whereas one gene showed the same expression level as the diploid strains. In both triploids, exceptional cases showed a negative correlation of expression with ploidy or a positive correlation greater than expected from the structural gene dosage. Interestingly, the transcription levels of most tested genes were signi®cantly dierent from the strains which were used as parents of the triploids, and also widely divergent expression patterns were found for some genes in the diploid ospring. In this study, the cause of the unexpected expression patterns observed in the euploid series is discussed in relation to the dierence between the two parental strains in expression level of genes and in trans-acting regulatory eects on their target genes. Keywords: Bombyx mori, gene dosage compensation, gene dosage eect, polyploidy, triploid silkworm. show approximately the same level of X chromosomelinked gene expression. Contrary to this, it has generally been regarded that the level of expression of an autosomal gene in higher eukaryotic organisms is directly proportional to the number of copies present in the genotype (Stewart & Merriam, 1974; Schimke et al., 1978; Birchler, 1983). This concept is known as gene dosage eect. However, a study in maize involving a dosage series of a whole chromosome arm showed dosage compensation rather than dosage eect for the alcohol dehydrogenase (Adh) locus (Birchler, 1979). Whereas compensation was observed for Adh, which was present in the varied segment, the level of expression of enzymes encoded elsewhere in the genome was inversely aected by the dosage series. The compensation in the whole arm series was shown to be caused by a cancellation of a structural gene dosage eect by a simultaneously produced inverse eect (Birchler, 1981). A similar phenomenon has been observed in Drosophila. Devlin et al. (1982) found that three of the ®ve loci
Introduction In many species, females have twice as many X chromosomes as males. This imbalance in gene content is recti®ed by dosage compensation, a regulatory mechanism that equalizes gene expression between individuals possessing one or two X chromosomes (Lucchesi, 1978). In mammals, equality of X chromosome-linked gene expression occurs by inactivation of one X chromosome in females through chromosome condensation. In contrast, dosage compensation in Drosophila is not mediated by chromosome condensation; instead, the transcriptional activity of X chromosome-linked genes is regulated to synthesize the same amount of gene product in both males and females. Thus, normal males (XY), normal females (XX) and metafemales (XXX) *Correspondence. E-mail:
[email protected] àPresent address: Laboratory of Molecular Entomology and Baculovirology, Institute of Physical and Chemical Research (RIKEN), Hirosawa 2±1, Wako, Saitama 351±0198, Japan.
Ó 1999 The Genetical Society of Great Britain.
661
662
M. G. SUZUKI ET AL.
monitored in larvae trisomic for the left arm of chromosome 2 exhibited similar levels of total expression relative to diploid controls. This phenomenon of autosomal dosage compensation has also been observed in larvae trisomic for the left arm of chromosome 3 (Devlin et al., 1985) and operates at the level of transcription (Devlin et al., 1984). In addition, when the dosage of a chromosomal segment is varied, trans-acting eects, either positive (direct) or negative (inverse), on genes elsewhere in the genome have been observed. Studies on expression patterns of six genes in a dosage series on 14 chromosomal segments in maize revealed that transcription levels of each tested gene were aected by multiple chromosomal regions (Guo & Birchler, 1994). Because of the predominant lethality of ploidy changes in animal systems, most of the studies on dosage regulation of ploidy series have been reported in yeast, Drosophila and animal cell lines (Ciferri et al., 1969; Priest & Priest, 1969; Lucchesi & Rawls, 1973). Among isozymes and proteins examined in various plant species, some are positively or negatively aected by ploidy changes, athough most of them exhibit an increased level as the ploidy level rises, that is, a gene dosage eect (DeMaggio & Lamb3 rukos, 1974; Levin et al., 1979; Timko et al., 1980). 1 In Bombyx mori, polyploids are easily induced when newly laid eggs are exposed to high or low temperatures or centrifugal force, or to chemicals such as colchicine (Tanaka & Kawaguchi, 1932; Hashimoto, 1933; Kawaguchi, 1934; Hirobe, 1939; Astaurov, 1967; Tazima & Onuma, 1967; Tamazawa & Takizawa, 1977). Therefore, this insect should be good material to help in understanding the dosage regulation of ploidy series in animal systems. The purpose of this investigation was to test whether aneuploid eects are cumulative, in principle, or whether dosage compensation occurs, as suggested by the previous observations in the maize and Drosophila trisomies. In this paper, we investigated gene dosage eects by varying the ploidy, where genomic balance is maintained as compared with aneuploids. The transcription levels of six genes expressed in various tissues were analysed in diploid and triploid silkworms. Contrary to the case in the maize ploidy series, in which most genes exhibited a gene dosage eect relative to the ploidy level (Guo et al., 1996), more than half of the tested genes in Bombyx showed a disproportionate increase or decrease in expression with ploidy level.
Materials and methods Genetic stocks and production of a ploidy series In Bombyx mori, many methods to induce triploids have been established (Tanaka & Kawaguchi, 1932; Hashi-
moto, 1933; Kawaguchi, 1934; Hirobe, 1939; Astaurov, 1967; Tazima & Onuma, 1967; Tamazawa & Takizawa, 1977; Yokoyama et al., 1990). In the present experiment, the triploid silkworms were obtained from eggs by heat-shock treatment as described in Yokoyama et al. (1990), which causes the fusion of an ameiotic female pronucleus (2n) with a male pronucleus (n). Observations of the chromosomes in embryonic cells and the phenotypic expression of genetic markers con®rmed that more than 90% of individuals produced by this method were triploids (Yokoyama et al., 1990). Control larvae were derived from eggs without the treatment. The diploid and the triploid obtained from eggs deposited by p50 female moths which were mated with C108 males are referred to as PC and PPC, respectively. The diploid and the triploid obtained from eggs resulting from a cross between C108 females and p50 males are denoted as CP and CCP, respectively. The eggs deposited by PPC moths and CCP moths, which were a mixture of abnormal and normal shapes, died before hatching. This situation was the same as that observed in the typical ZZW triploid individuals reported by Sato & Chino (1937) and Yokoyama et al. (1990). Silkworm strains p50 and C108 were originally supplied by the Institute of Genetic Resources, Faculty of Agriculture, Kyusyu University, Fukuoka, Japan, and the National Institute of Genetics, Mishima, Japan, respectively. Larvae were reared on fresh mulberry leaves. The midgut, middle silk gland and fat body were dissected out in PBS (137 mM NaCl, 2.7 mM KCl, 4.3 mM Na2HPO4, 1.4 mM KH2PO4) from day-2 ®fth instar larvae, frozen in liquid nitrogen, and stored at ±80°C. Ovaries were obtained from day-5 pupae. Probe preparation The transcription level of the following genes was analysed: Storage protein 1 (23±8.6), Storage protein 2 (3±16.7), Alkaline phosphatase (3±unknown), Trehalase (unknown), Bombyx Antennapedia (6±0.0) and chorion gene AL12 (2±6.9) (the map position of each gene is in parentheses.) Storage protein 1 and 2 genes were obtained from cDNA plasmids pBmSPIC and pBmSP2C2 provided by Dr S. Izumi, Tokyo Metropolitan University, Tokyo, Japan. The Alkaline phosphatase gene was obtained from a clone used in a previous study (Suzuki et al., 1998). The trehalase gene was synthesized by PCR ampli®cation with primers TEF1 (5¢-TCCGACCAGTTTACTGCAACA-3¢) and TER2 (5¢-TTGCAGTTTCGGTCTGTCAG-3¢), which were designed using the cDNA sequence (Su et al., 1994), and cloned into the TA cloning site of the pGEM-T vector (Promega, Madison, WI). The Bombyx Ó The Genetical Society of Great Britain, Heredity, 82, 661±667.
GENE EXPRESSION IN THE TRIPLOID SILKWORM
Antennapedia gene was derived from a cDNA clone, supplied by Dr K. Ueno, National Institute of Basic Biology, Okazaki, Japan. The inserts of these clones were excised with appropriate restriction ends nucleases, and then labelled with 32P-dCTP using a Random Primer DNA Labelling Kit Version 2 (Takara Shuzo, Kyoto, Japan) according to the manufacturer's instructions. Northern analysis Total RNA was isolated by the acid guanidinium thiocyanate phenol-chloroform (AGPC) method 14 (Chomczynski & Sacchi, 1987). RNA was electrophoresed on 1.2% agarose±2.2 M formamide gels in MOPS buer and transferred onto a Hybond-N+ nylon membrane (Amersham International plc, Buckinghamshire, UK). The ®lter was then hybridized and washed following the procedures described previously (Suzuki et al., 1998). For each gene, ®ve loadings of each dosage series were made on each gel. After probing with each speci®c 15 gene, the blots were reprobed with ribosomal DNA as a measure of the loading of total RNA. To test that rRNA is a valid loading control, i.e. that a constant rRNA level is produced per genome among the ploidy series, total nucleic acid isolated from midgut, middle silk gland and fat body were separated on an agarose gel and stained with ethidium bromide. The DNA and rRNA bands were scanned with a pdi 420 oe scanner (pdi, Huntington Station, NY). DNA:rRNA ratios were then calculated using the electrophoretic band analysis software DIVERSITY ONE version 1.5 (pdi, Huntington Station, NY). Data analysis Radioactivity was visualized and quanti®ed using a BAS 2000 Bioimage Analyser (Fuji Photo Film, Tokyo, Japan). The speci®c RNA quantity of each lane was normalized by calculating the ratio of mRNA/rRNA using the BAS 2000 readings. The means and standard errors of the respective normalized values from each ploidy were calculated from ®ve replicates. The mRNA levels of the triploid relative to that of the diploid were determined by taking the ratios of the respective mean values of ®ve replicates. Such ratios represent mRNA levels relative to those of the diploid on a per genome basis because rRNA/DNA ratios were constant among the ploidy series. The values on a per cell basis were calculated by multiplying the per genome values by the respective genomic ratios of each ploidy to the diploid: i.e. 1.0 for diploid, 1.5 for triploid. Ó The Genetical Society of Great Britain, Heredity, 82, 660±667.
663
Results Equal ribosomal RNA levels were present per genome between the diploid and the triploid To con®rm that ribosomal RNA could be used to standardize loading of RNA on Northern blots, total nucleic acids from the fat body, midgut and middle silk gland of the diploid and triploid were quanti®ed and compared. The ratio of rRNA/DNA was calculated, but no evidence of variation was found between the triploid and the control diploid (Table 1, compare PPC and PC, CCP and CP). Therefore, rRNA could be used as a valid internal loading control. Comparison of the transcription levels of six genes in the diploid To determine whether gene dosage eects are additive or nonadditive in general when the ploidy is varied, Northern analysis was used to determine the transcription levels of six dierent genes. Values from the Northern analysis (Table 2) are expressed as the relative transcription levels of each gene from the triploid to that of the diploid and are normalized on a per genome basis. If a gene is expressed in proportion to its dosage, an equal expression value will be measured per genome regardless of the ploidy level. Results from the study indicate that the transcription levels of the Storage protein-2 (SP-2: chromosome 3) gene are approximately equal among the strains C108 and p50, which were used as parents to produce the triploid, and the control diploid, PC and CP (Tables 2 and 3). But the transcription levels of the Storage protein-1 gene (SP-1: chromosome 23), the Alkaline phosphatase gene (Alp: chromosome 3) and the AL-11 gene (AL-11: chromosome 2) were signi®cantly dierent
Table 1 Comparison of the rRNA/DNA ratio between the diploid and the triploid in Bombyx mori Fat body Ploidy Mean SEà PPC PC CCP CP
3n 2n 3n 2n
8.81 8.08 7.10 7.21
1.61 0.90 0.71 1.31
Middle silk gland
Posterior silk gland
Mean
SE
Mean
SE
3.86 3.37 3.08 3.11
0.34 0.11 0.24 0.24
14.74 13.16 11.83 11.75
2.09 1.86 2.83 1.66
Mean is an average value of the rRNA/DNA ratio among ®ve individuals. None of the comparisons between the diploid and the triploid data is statistically dierent at the 5% signi®cance level. àSE indicates the standard error of ®ve replicates.
664
M. G. SUZUKI ET AL.
Table 2 The mRNA level of six Bombyx mori genes on a per genome basis in the triploid relative to the diploid SP-1 (23 )
PPC PC CCP CP C108 p50
SP-2 (3)
Alp (3)
Tre (unknown)
Bm Antp (6)
AL-11(2)
Ploidy
Meanà
SE§
Mean
SE
Mean
SE
Mean
SE
Mean
SE
Mean
SE
3n 2n 3n 2n 2n 2n
0.77 1.03 1.47 1.00 1.63 0.51
0.03 0.09 0.02 0.03 0.12 0.07
1.20 0.99 1.28 1.00 1.08 1.06
0.13 0.06 0.12 0.07 0.06 0.03
0.77 1.02 1.14 1.00 1.33 0.61
0.05 0.07 0.07 0.07 0.04 0.04
0.60 1.03 0.98 1.00 0.69 0.98
0.02 0.05 0.02 0.04 0.02 0.05
1.44 1.03 0.81 1.00 0.72 1.06
0.11 0.06 0.08 0.06 0.14 0.03
0.96 0.92 0.67 1.00 0.18 0.95
0.04 0.02 0.08 0.12 0.06 0.03
Numbers in parentheses represent the chromosome where each gene is mapped. àMean is the average value of the relative mRNA levels from ®ve replicates. Relative level shows the ratio of the mRNA level obtained to the amount resulting from the diploid ospring, CP (1.00). §SE indicates the standard error of ®ve replicates.
between C108 and p50. The transcription levels of SP-1 and Alp have 3.2-fold and 2.2-fold increases in C108 as compared with those of p50, respectively. In contrast, AL-11 had a 5.2-fold higher mRNA level in p50 as compared with C108. Although less dramatic, there was a clear dierence in the mRNA levels of the remaining genes (the trehalase gene (Tre: unknown) and the Bombyx Antennapedia gene (Bm Antp: chromosome 6)) between C108 and p50. The transcription level of Tre had a 1.42-fold increase in p50 compared with C108; Bm Antp had a 1.46-fold higher mRNA level in p50 compared with C108 (Tables 2 and 3). On the other hand, the mRNA levels of all genes in PC and CP were approximately the same. For SP-2, the transcription levels in PC and CP were equal to those in C108 and p50. The mRNA levels of SP-1 and Alp in PC and CP were almost the same as the mid-value of those in C108 and p50. Interestingly, the transcription levels of Tre, Bm Antp and AL-11 in PC and CP were identical to the higher mRNA level observed in the parents, i.e. the level displayed in p50. These results indicate that whereas the mRNA levels of many genes in ospring may show the
mid-value of the transcription level in the parental strains, some genes display exceptional mRNA levels, including Tre, Bm Antp and AL-11. Comparison of the transcription level of each gene between the diploid and triploid In the triploid ospring PPC, the transcription level per cell of two genes, SP-2 and AL-11, increased proportionally with increasing ploidy, as observed in the previous report on dosage eects in a maize ploidy series. In contrast to such standard dosage eects, the remaining genes showed unexpected levels of mRNA. The mRNA level of Bm Antp was signi®cantly higher than expected, on a per cell level; taking the diploid ospring PC as 100% (the approximate mid-parental value), the value of the PPC was 210%. There was also a slight increase in the transcription levels of SP-1 and Alp from diploid to triploid. The increased mRNA levels of these gene per cell values with increasing ploidy were 113% (SP-1) and 114% (Alp), respectively (Table 3).
Table 3 The mRNA level of six Bombyx mori genes on a per cell basis in the triploid relative to the diploid SP-1 (23 )
PPC PC CCP CP C108 p50
SP-2 (3)
Alp (3)
Tre (unknown)
Bm Antp (6)
AL-11(2)
Ploidy
Meanà
SE§
Mean
SE
Mean
SE
Mean
SE
Mean
SE
Mean
SE
3n 2n 3n 2n 2n 2n
2.32 2.05 4.40 2.00 3.25 1.01
0.10 0.19 0.06 0.00 0.24 0.15
3.58 1.97 3.83 2.00 2.15 2.12
0.39 0.14 0.37 0.14 0.14 0.05
2.31 2.05 3.47 2.00 2.70 1.25
0.16 0.13 0.23 0.13 0.09 0.06
1.81 2.05 2.97 2.00 1.37 1.96
0.04 0.08 0.05 0.08 0.04 0.09
4.33 2.06 2.42 2.00 1.44 2.11
0.33 0.11 0.25 0.11 0.28 0.06
2.87 1.85 2.00 2.00 0.37 1.90
0.12 0.03 0.23 0.23 0.11 0.06
Numbers in parentheses represent the chromosome where each gene is mapped. àMean is an average value of the relative mRNA levels from ®ve replicates. Mean values that are in bold face are signi®cantly dierent (P < 0.01 in t-tests) from those of diploids. Mean values that are not statistically dierent from the dose-dependent levels at the 5% signi®cance level are in italics. §SE indicates the standard error of ®ve replicates.
Ó The Genetical Society of Great Britain, Heredity, 82, 661±667.
GENE EXPRESSION IN THE TRIPLOID SILKWORM
Another exceptional pattern was exhibited by Tre. The transcription level of Tre on a per cell basis slightly decreased as the ploidy increased (Table 3), to 88% of the diploid level. On the other hand, in the CCP triploids, the mRNA levels per cell of three genes, SP-2, Alp and Tre, were approximately as expected if no compensation occurred. In contrast, the mRNA level of AL-11 was almost the same as that in the diploid ospring CP, indicating that the additional gene dose was compensated. Exceptional cases were also seen in CCP. The mRNA level of Bm Antp in CCP was »21% greater than that in CP, showing a weak over-compensatory eect. The transcription of SP-1 was signi®cantly higher than expected. The mRNA level of this gene increased by 220% in comparison to the diploid. From these results, it appeared that the only gene that showed the standard dosage eect in both PPC and CCP was SP-2. All of the other genes showed dierent expression patterns between PPC and CCP. In contrast to what was found in the maize ploidy series, where the expression of most genes was dosage dependent, more than half of the genes displayed unexpected levels of mRNA in the present study.
Discussion Polyploids in numerous plant species have shown higher levels of enzyme activity of proteins than diploids (DeMaggio & Lambrukos, 1974; Nakai, 1977; Levin et al., 1979; Timko et al., 1980). The study on the maize ploidy series demonstrated that increasing gene dosage from one to four via a ploidy series increases the expression of most genes proportionally (Guo et al., 1996). Similar observations were also made with yeast (Ciferri et al., 1969) and Drosophila (Lucchesi & Rawls, 1973). Similarly, in the ®broblast-like cell lines of rat, the rate of collagen synthesis was correlated with ploidy (Priest & Priest, 1969). Although exceptions exist, such direct correlations of gene expression with the copy number of the genome in a ploidy series are the predominant situation. In contrast, results from this study indicated that more than half of six genes with widely diering function displayed exceptional transcription levels. SP-2 was the only gene which showed the standard dosage eect and approximately the same mRNA level in both the PPC and CCP triploids. This gene was also the only gene for which the mRNA level was almost the same among diploid parental strains and their diploid ospring, PC and CP. These ®ndings suggest that the transcription levels of genes in aneuploids may show dose dependence when those in diploid parental strains are equal to each other. Ó The Genetical Society of Great Britain, Heredity, 82, 660±667.
665
The predominant tendency that genes duplicated in triploids show strict dose dependence, strongly supports a prediction that the expected transcription level would simply represent the sum of the mRNA levels of each of the three gene copies. Therefore, the mRNA levels in the triploid ospring PPC and CCP were calculated as (the copy number of a gene derived from p50) ´ (its mRNA level in p50 ¸ 2) + (the number of a gene derived from C108) ´ (its mRNA in C108 ¸ 2). For example, the mRNA levels of SP-1 in PPC and CCP are calculated as the following. PPC: 2 ´ (1.01/2) + 1 ´ (3.25/2) 2.64; CCP: 1 ´ (1.01/2) + 2 ´ (3.25/2) 3.76. The observed and calculated transcription levels are presented in Table 4. The mRNA levels of three genes, SP-1, SP-2 and Alp, are approximately identical to the calculated levels, indicating that they show dose dependence. From these ®ndings, it is apparent that genes duplicated in the triploid ospring exhibit exceptional transcription levels in spite of their dose-dependent expression when there is a dierence between the mRNA levels in each of the diploid parental strains. As that dierence increases, the degree of deviation from expected mRNA level in the triploid (i.e. 1.5-fold higher mRNA level as compared with the diploid ospring) becomes extreme (Table 4). Contrary to the results for these genes, Tre, Bm Antp and AL-11 do not follow this trend and show quite remarkable expression patterns. The transcription level of Tre in PPC was slightly below the diploid level, despite an increase in the copy number of the genome. The mRNA level of Bm Antp showed a 1.59-fold increase in PPC and a 1.44-fold decrease in CCP compared with the calculated levels. The mRNA level of AL-11 increased 1.38-fold in PPC as compared with the calculated level, which was almost the same as the dose-dependent level. In CCP, AL-11 mRNA level was 1.52-fold greater than the calculated level; however, this level was identical with the diploid level. These genes also showed unexpected transcription levels in the diploid ospring. The transcription levels of all of these genes were approximately identical to the level displayed in p50; in other words, the mRNA levels 16 of these genes were equal to the higher of the mRNA levels observed in the parental strains, whereas others showed levels equal to the sum of the transcription level per allele in each of the parental strains. This observation supports the possibility that C108 alleles of these genes in the PC and CP diploids are activated by transacting regulatory genes contained in the p50 genome. This causes an increase of the mRNA level of the C108 alleles to the same levels as those displayed by the p50 alleles. As a result, the transcription levels of these genes in PC and CP would reach the same levels as those in strain p50. Tre, Bm Antp and AL-11 may show remarkable expression patterns in PPC and CCP
666
M. G. SUZUKI ET AL.
Table 4 Comparison of mRNA levels of six Bombyx mori genes between the observed and the estimated data SP-1 (23 )
SP-2 (3)
Ploidy Observed dataà Estimated data§ PPC PC CCP CP
3n 2n 3n 2n
2.32 2.05 4.40 2.00
0.10 0.19 0.06 0.00
2.64 2.13 3.76 2.13
0.27 0.20 0.32 0.20
Observed data 3.58 1.97 3.83 2.00
Tre (unknown) Ploidy PPC PC CCP CP
3n 2n 3n 2n
Observed data 1.81 2.05 2.97 2.00
0.04 0.08 0.04 0.08
Estimated data 3.20 2.14 3.21 2.14
0.12 0.10 0.17 0.10
Observed data Estimated data 2.31 2.05 3.47 2.00
Bm Antp (6)
Estimated data 2.65 1.67 2.35 1.67
0.39 0.14 0.37 0.14
Alp (3)
0.11 0.07 0.09 0.07
Observed data 4.33 2.06 2.42 2.00
0.33 0.11 0.25 0.11
2.60 1.98 3.33 1.98
0.11 0.08 0.12 0.08
AL-11 (2)
Estimated data 2.83 1.78 2.50 1.78
0.16 0.13 0.23 0.13
0.20 0.17 0.31 0.17
Observed data Estimated data 2.87 1.85 2.00 2.00
0.12 0.03 0.23 0.23
2.09 1.14 1.32 1.14
0.12 0.09 0.14 0.09
Numbers in parentheses represent the chromosome where each gene is mapped. àObserved data represent a `mean' `SE', each of which is shown in Table 3. Values that are not statistically dierent from the estimated data at the 5% signi®cance level are in bold face. §Estimated data are a result of the calculation using the following equation: (the number of a gene derived from p50) ´ (the mRNA level of a gene in p50/2) + (the number of a gene derived from C108) ´ (the mRNA level of a gene in C108/2).
because trans-acting regulatory eects also act on these genes, in addition to gene dosage eects. According to our ®ndings, the unexpected expression patterns which have been observed in the previous reports on ploidy series may result from the following two reasons: (i) the expression levels of genes exhibit dierences between the two parental strains; (ii) there is a dierence between structural genes of the two parental strains with respect to the degree to which they are enhanced or inhibited by their trans-acting regulators. When this degree is identical with the parental strains, structural genes display a gene dosage eect in a ploidy series. To con®rm the validity of our conclusion, further work is required to determine whether or not genes which showed unexpected expression levels in the present study, such as Tre, Bm Antp and AL-11, display a strict gene dosage eect in the euploid series derived from a single strain.
Acknowledgements We wish to thank Dr Marian R. Goldsmith, University of Rhode Island, for her helpful advice and reviewing this manuscript. This work was supported in part by Grants-in-Aid for Scienti®c Research from the Ministry of Education, Science, Sports and Culture, Japan (no. 08276101 to T. S. and 09306004 to M. K.) and by CREST (Core Research for Evolutional Science and Technology) of the Japan Science and Technology Corporation (to T. S.).
References ASTAUROV, B. L.
1967. Experimental alteration of the developmental cytogenetic mechanisms in mulberry silkworms: arti®cial parthenogenesis, polyploidy, gynogenesis and androgenesis. Adv. Morph., 6, 199±257. BIRCHLER, J. A. 1979. A study of enzyme activities in a dosage series of the long arm of chromosome one in maize. Genetics, 92, 1211±1229. BIRCHLER, J. A. 1981. The genetic basis of dosage compensation of alcohol dehydrogenase-1 in maize. Genetics, 97, 625±637. BIRCHLER, J. A. 1983. Allozymes in gene dosage studies. In: Tanksley, S. D. AND Orton, T. J. (eds) Isozymes in Plant Genetics and Breeding, pp. 85±108. Elsevier, Amsterdam. CHOMCZYNSKI, P. AND SACCHI, N. 1987. Single-step method of RNA isolation by acid guanidinium thiocyanate-phenolchloroform extraction. Analyt. Biochem., 162, 156±159. CIFERRI, O., SORA, S. AND TIBONI, O. 1969. Eects of gene dosage on tryptophan synthase activity in Saccharomyces cerevisiae. Genetics, 61, 567±576. DEMAGGIO, A. E. AND LAMBRUKOS, J. 1974. Polyploidy and gene dosage eects on peroxidase activity in ferns. Biochem. Genet., 12, 429±440. DEVLIN, R. H., HOLM, D. G. AND GRIGLIATTI, T. A. 1982. Autosomal dosage compensation in Drosophila melanogaster strains trisomic for the left arm of chromosome 2. Proc. Natl. Acad. Sci. U.S.A., 79, 1200±1204. DEVLIN, R. H., GRIGLIATTI, T. A. AND HOLM, D. G. 1984. Dosage compensation is transcriptionally regulated in autosomal trisomics of Drosophila. Chromosoma, 91, 65±73. DEVLIN, R. H., HOLM, D. G. AND GRIGLIATTI, T. A. 1985. Gene dosage compensation in trisomics of Drosophila melanogaster. Devl. Genet., 6, 39±58. Ó The Genetical Society of Great Britain, Heredity, 82, 661±667.
GENE EXPRESSION IN THE TRIPLOID SILKWORM GUO, M. AND BIRCHLER, J. B.
1994. Trans-acting dosage eects on the expression of model gene systems in maize aneuploids. Science, 266, 1999±2002. GUO, M., DAVIS, D. AND BIRCHLER, J. A. 1996. Dosage eects on gene expression in a maize ploidy series. Genetics, 142, 1349±1355. HASHIMOTO, H. 1933. Genetical studies on the tetraploid female in the silkworm. Bull. Seric. Exp. Sta. Jap., 8, 359±381. HIROBE, T. 1939. Polyploid silkworm induced by colchicine treatment upon eggs. Jap. J. Genet., 15, 69±74. KAWAGUCHI, E. 1934. Chromosome behavior in tetraploid female of the silkworm. J. Seric. Sci. Jap., 5, 93±94. LEVIN, D. A., TORRES, A. M. AND LEVY, M. 1979. Alcohol dehydrogenase activity in diploid and autotetraploid Phlox. Biochem. Genet., 17, 35±43. LUCCHESI, J. C. 1978. Gene dosage compensation and the evolution of sex chromosomes. Science, 202, 711±716. LUCCHESI, J. C. AND RAWLS, J. M. 1973. Regulation of gene function: a comparison of enzyme activity levels in relation to gene dosage in diploids and triploids of Drosophila melanogaster. Biochem. Genet., 9, 41±51. NAKAI, Y. 1977. Variations of esterase isozymes and some soluble proteins in diploids and their induced autotetraploids in plants. Jap. J. Genet., 52, 171±181. PRIEST, R. E. AND PRIEST, J. H. 1969. Diploid and tetraploid clonal cells in culture: gene ploidy and synthesis of collagen. Biochem. Genet., 3, 371±382. SATO, S. AND CHINO, K. 1937. Biological measurement of the triploid silkworm, Bombyx Mori. J. Seric. Sci. Jap., 8, 107±120. SCHIMKE, R. T., KAUFMAN, R. F., ALT, F. W. AND KELLEMS, R. F. 1978. Gene ampli®cation and drug resistance in cultured murine cells. Science, 202, 1051±1055.
Ó The Genetical Society of Great Britain, Heredity, 82, 660±667.
STEWART, R. T. AND MERRIAM, J. R.
667
1974. Segmental aneuploidy and enzyme activity as a method for cytogenetic localization in Drosophila melanogaster. Genetics, 76, 301±309. SU, Z.-H., IKEDA, M., SATO, Y., SAITO, H., IMAI, K., ISOBE, M. AND YAMASHITA, O. 1994. Molecular characterization of ovary trehalase of the silkworm, Bombyx mori and its transcriptional activation by diapause hormone. Biochim. Biophys. 7 Acta, 1218, 366±374. 1 SUZUKI, G. M., SHIMADA, T. AND KOBAYASHI, M. 1998. Absence of dosage compensation at the transcription level of a sexlinked gene in a female heterogametic insect, Bombyx mori. Heredity, 81, 275±283. TAMAZAWA, S. AND TAKIZAWA, Y. 1977. On the polyploid induced by super cooling treatment of the eggs of the silkworm, Bombyx mori L. The relation between the enlargement of the serosa cells and the polyploidy. Mem. Fac. Agric. Hokkaido University, 10, 272±283. TANAKA, Y. AND KAWAGUCHI, E. 1932. On the triploid silkworm produced by the centrifugal force treatment. Jap. J. Genet., 7, 186±187. TAZIMA, Y. AND ONUMA, A. 1967. Experimental induction of androgenesis, gynogenesis, and polyploidy in Bombyx mori by treatment with CO2 gas. J. Seric. Sci. Jap., 36, 286±292. TIMKO, M. P., VASCONCELOS, A. C. AND FAIRBROTHERS, D. E. 1980. Euploidy in Ricinus. I. Euploidy and gene dosage eects on cellular proteins. Biochem. Genet., 18, 171±193. YOKOYAMA, T., SUGAI, E., OSHIKI, T. AND PAN, Q. 1990. Induction of the triploid silkworm, Bombyx mori, by the hot-water treatment to the inseminated eggs immediately after oviposition. J. Seric. Sci. Jap., 59, 281±224.
