Glucose Dehydrogenases in Drosophila Melanogaster 6 - Phosphate and 6 - Phosphogluconate

By combining ten second and ten third chromosomes, we investigated chromosomal interaction with respect to the action of the modifier factors on GGPD and GPGD activities in Drosophila melanogaster. Analysis of variance revealed that highly significant chromosomal interaction exists for both enzyme activities. From the estimated variance components, it was concluded that the variation in enzyme activity attributed to the interaction is as great as the variation attributed to the second chromosome but less than attributed to the third chromosome. The interaction is not explained by the variation of body size (live weight). The interaction is generated from both the lack of correlation of second chromosomes for third chromosome backgrounds and the heterogeneous variance of second chromosomes for different third chromosome backgrounds. Large and constant correlation between GGPD and GPGD activities were found for third chromosomes with any second chromosome background, whereas the correlations for second chromosomes were much smaller and varied considerably with the third chromosome background. This result suggests that the activity modifiers on the second chromosome are under the influence of third chromosome factors. HE two oxidative pentose phosphate pathway enzymes of D. melanogaster, T glucose 6-phosphate dehydrogenase (GGPD, EC 1.1.1.49) and 6-phosphogluconate dehydrogenase (GPGD, EC 1.1.1.44), have been the subject of a large number of genetic, biochemical and physiological studies (see reviews by LUCCHESI, HUGHES and GEER 1979; GEER et al. 1981). The structural genes for both enzymes have been localized to the X chromosome, that for GPGD to the distal tip (Pgd at 1-0.6 and 2D3-5, YOUNG 1966; GERASIMOVA and ANAVIEV 1972) and that for G6PD to the proximal end (Zw at 1-63 and 17B18F, YOUNG, PORTER and CHILDS 1964; STEWART and MERRIAM 1974). Each locus is polymorphic for two common electrophoretic variants in natural populations (O’BRIEN and MACINTYRE 1969), and a number of null or low activity variants of each enzyme have been induced on the X chromosome (see LUCCHESI, HUGHES and GEER 1979). The active form of GPGD is a dimer (HORI and TANDA 1980; WILLIAMSON, KROCHKO and GEER 1980), whereas the natGenetics 106 655-668 April, 1984. 656 N. MIYASHITA AND C. C. LAURIE-AHLBERG ural polymorphism at Zw is due to instability of subunit association where one variant is a dimer and the other a tetramer (STEELE, YOUNG and CHILDS 1968; HORI and TANDA 1980). The mechanisms that regulate or cause variation in the activity levels of these enzymes have been investigated at several levels. Environmental causes include short-term fluctuations in the concentrations of metabolites that directly modulate activity levels as well as long-term influences of the diet that affect accumulation of enzyme molecules (GEER et al. 1981). The genetic causes of variation include sex-specific effects such as dosage compensation (LUCCHESI 1977; BELOTE and LUCCHESI 1980) and sex-nonspecific effects such as the difference in activity level between the A and B allozymes of 6PGD (BIJLSMA and VAN DER MEULEN-BRUIJNS 1979; CAVENER and CLEGG 1981; HORI and TANDA 1981) or the activity effects due to autosomal modifiers (HORI and TANDA 198 1 ; LAURIE-AHLBERG et al. 198 1). Genotype-environment interaction effects have been described in terms of the response of different genotypes to the modulation of activity levels by dietary carbohydrate (COCHRANE and LucWe have been concerned with the detection and characterization of naturally occurring genetic variants affecting the expression of G6PD and GPGD, with particular attention on the possibility of coordinate genetic control because of the closely related functions of these two enzymes (LAURIE-AHLBERG et al. 1980, 1981, 1982; WILTON et al. 1982). These experiments have utilized two sets of chromosome substitution lines with coisogenic backgrounds (50 second and 50 third chromosomes sampled at random from four different geographic localities). We find extreme modifier genetic effects due to each autosome, which are very repeatable over time and are generally substantially larger than the effects due to uncontrolled variation in the standard laboratory culture environment. Furthermore, the genetic effects on G6PD and 6PGD are highly correlated with each other as well as with some other metabolically related enzymes (WILTON et al. 1982). Tissue distribution studies shows that these activity effects are not restricted to one particular body part and may even go in opposite directions in different body parts. Immunoelectrophoresis experiments show that a large part (but perhaps not all) of the modifier variation is accounted for by variation in the concentration of enzyme molecules, especially for third chromosome lines (LAURIE-AHLBERG et al. 1981). Here, we extend our investigation of autosomal modifier effects on G6PD and 6PGD with a study of chromosomal interaction with respect to the activity of each enzyme individually and also with respect to the correlation between their activities. CHESI 1980). MATERIALS AND METHODS Line construction: Ten second and ten third chromosome substitution lines with coisogenic background were selected from a total of 50 of each type so as to represent the range of GGPD activity variation (see LAURIE-AHLBERG et al. 1980 for construction of these original lines). The goal was to establish all of 100 possible combinations between these second and third chromosomes, which were derived from natural populations. The original 20 lines have the same X chromosome from a highly inbred line Ho-R. Consequently, the combination lines have the X chromosome, which has alleles for the fast electrophoretic forms of both GGPD and GPGD. The procedure for conCHROMOSOMAL INTERACTION 657 structing lines homozygous for particular combinations of these chromosomes is shown in Figure 1. During this procedure females are never doubly heterozygous for second and third chromosome balancers with the wild-type chromosomes. Only seven of the possible 100 combinations failed because of synthetic lethality or weak expression of the marker genes. After establishment of the 93 combination lines, a starch gel electrophoretic survey of eight commonly polymorphic enzymes was conducted to check for errors in the procedure. No problems were detected. The electrophoretic procedures for the eight enzymes (ADH, EC 1 . 1 . 1 . 1 ; GPDH, EC 1.1.1.8; ODH, EC 1.1.1.73; PGM, EC 2.7.5.5; EST-6, EC 3.1.1.1; EST-C, EC 3.1.1.2; G6PD; GPGD) are described by LAURIEAHLBERG and WEIR (1 979). Enzyme assay: A sample of ten males was homogenized in 0.5 ml of 0.01 M potassium phosphate buffer, pH 7.4, with 1 mM EDTA, 5 mM DTT and 0.5% (v/v) Triton X-100 and centrifuged for 10 minutes at 10,000 X g. The supernatant was used for activity measurements and the determiil i 2 +3 +I T ( 2 ; 3 ) a p X a X x il Rn TM6 Y i 2 i 3 , 11 i 2 +3 , Y CyO;TM6 7 ii i2 +3 11 Cy0 TM6 r-X x ii Rn TM6 y i 2 +3 , 11 i 2 +3 , Y i 2 +3 ii cy0 4-3 ii c > T 4 6 x t-il Pm +3 , Y +2 i 3 I I il 5 2 5 3 '1 +2 +3 il +2 +3 Y +2 +3 FIGURE 1.-Procedure for construction of a line homozygous for both a second and a third chromosome. i = chromosome from Ho-R inbred line; + = chromosome from natural population; Cy0 = Zn(2LR)O, Cy; Pm = In(2LR)bw"'; TM6 = Zn(3LR)Tm6, Ubx. 658 N. MIYASHITA AND C. C. LAURIE-AHLBERG nation of total amount of protein. Enzyme activities were measured by observing the reduction of NADP+ to NADPH at 340 nm. The reaction mixture contains 0.1 ml of the supernatant and 0.9 ml of 0.055 M Tris-HCI buffer, pH 7.6, with 18.5 mM MgClp, 0.18 mM NADP+ and 1.8 mM glucose-6-phosphate for G6PD activity and 0.9 ml of 0.055 M Tris-HCI buffer, pH 7.6, with 1.68 mM MgS04, 0.15 mM NADP+ and 0.34 mM 6-phosphogluconate for GPGD activity. The crude supernatant was diluted in a 1:9 ratio with distilled water, and then the amount of protein was determined by the method of LOWRY et al. (1 95 1). Experimental deszgn: The design of this experiment is shown in Figure 2. The combination lines were arranged in rows and columns in a random fashion according to the origins of the chromosomes. Then, the combination lines were split into four groups (Hi Vj, i j = 1,2) in order to accommodate the number of assays that could be performed in 1 day. Two replicates of ten 4day-old males were sampled from each line in each group. The samples were weighed and kept frozen at -70" until the assay of enzyme activities. The samples from one group were assayed in 1 day for both G6PD and GPGD activities. The sampling and assay for each of the groups was repeated four times (four blocks). This design yields a total of eight observations per line. However, some samples were lost during the experiment due to low viability and/or fertility. Thus, a total of 728 samples was assayed. Flies were raised in the standard cornmeal-molasses medium at 25". The model for the analysis of variance for each of the variables, GGPD, GPGD, weight and