THE c~-GLYCEROPHOSPHATE CYCLE IN DROSOPHILA MELANOGASTER IV. Metabolic, Ultrastructural, and Adaptive Consequences of aGpdh-1 "Null" Mutations

"Nul l" mutations previously isolated at the aGpdh-1 locus of Drosophila rnelanogaster, because of disruption of the energy-producing a-glycerophosphate cycle, severely restrict the flight ability and relative viability of affected individuals. Two "null" alleles, aGpdh-18°-1-4, and otGpdh-1 ~°-~-5, when made hemizygous with a deficiency of the aGpdh-I locus, Df(2L)Gdh,4, were rendered homozygous by recombination with and selective elimination of the Dfl2L)GdhA chromosome. After over 25 generations, a homozygous aGpdh-1 ~°-1-4 stock regained the ability to fly despite the continued absence of measurable a G P D H activity. Inter se heterozygotes of three noncomplementing aGpdh-1 "null" alleles and the "adapted" aGpdh-1 B°-'-~ homozygotes were examined for metabolic enzymatic activities related to the energy-producing and pyridine nucleotide-regulating functions of the o~-glycerophosphate cycle in Drosophila. The enzyme functions tested included glyceraldehyde-3-phosphate dehydrogenase, cytoplasmic and soluble malate dehydrogenase, lactate dehydrogenase, mitochondrial N A D H oxidation, oxidative phosphorylation, and respiratory control with the substrates a-glycerophosphate, succinate, and pyruvate. These activities in any of the mutant genotypes in early adult life were indistinguishable from those in the wild type. There was, however, a premature deterioration and atrophy of the ultrastructural integrity of flight muscle sarcosomes observed by electron microscopy in the "null" mutants. These observations were correlated with a decrease in state 3 mitochondrial oxidation with a-glycerophosphate, succinate, and pyruvate, as well as with loss of respiratory control in adults as early as 2 wk after eclosion. Such observations, which normally are seen in aged dipterans, were accompanied by premature mortality of the mutant heterozygotes. The adapted aGpdh-I B°-14 was identical with wild type in each of the aging characters with the single exception of lowered rates of mitochondrial oxidative phosphorylation. 8 6 4 THE JOURNAL OF CELL BIOLOGY VOLUME 63, 1974 • p a g e s 8 6 4 8 8 2 on D ecem er 3, 2017 jcb.rress.org D ow nladed fom The evolutionary requirement for both genic variability and recombinational plasticity provides a selective pressure on biological species in favor of mechanisms which preserve the temporary accumulation of mutational and recombinational variability. This selective pressure has produced and preserved certain biological "buffer systems" which protect spontaneous mutations and randomly produced gene combinations from the immediate rigors of natural selection (1, 2). These buffers include diploidy itself, functional redundancy (e.g., two biochemical routes to the same product), and retention of vestigal functions during ontogeny. A remarkable demonstration of such buffering is emphasized by the observation that of the 14 genes in Drosophila melanogaster at which "null" or "silent" alleles eliminate their respective gene products, only one, bobbed, has lethal alleles (3). "'Null" alleles at each of the other geneenzyme loci are all homozygous viable, despite the absence of presumably important functions. A particularly important function which exhibits such protection is the a-glycerophosphate (aGP) cycle, a major circuit in the energy-producing machinery of the insect flight muscle (4-8). The ctGP cycle consists of two biochemically and genetically distinct a-glycerophosphate dehydrogenases: a cytoplasmic NAD like a-glycerophosphate dehydrogenase (aGPDH; Lglycerol 3-phosphate:NAD oxidoreductase, EC 1.1.1.8), and a particulate mitochondrial inner membrane flavoprotein, a-glycerophosphate oxidase (a-GPO; L-glycerol 3-phosphate:cytochrome c oxidoreductase, EC 1.1.99.5). The otGP cycle provides an efficient means to effect three important functions: (a) maintenance of NAD-NADH equilibrium in the cytoplasm, (b) ATP production for muscular contraction, and (c) provision of aGP for phospholipid anabolism (4 6). The two enzymes in D. melanogaster are products of different genes. The structural gene for ctGPDH, designated aGpdh-1, is located on 2L at 2-20.5 (7, 9); aGPO is not affected by mutations at this locus but is sensitive to aneuploidy in a gene dosage response manner at a region (50C-52E) on 2R (10). This region probably contains the structural gene, although direct evidence is lacking. Despite the genetic distinctions, certain biochemical and developmental correlations between pairs of isozymes of the two enzymes suggest metabolic cooperation in the operation of the , G P cycle (6). Four "null" or silent mutant alleles at the aGpdh-1 locus were previously isolated after mutagenesis with ethyl methane sulfonate (EMS) (7). Flies heterozygous for any of the noncomplementing alleles or hemizygous with a deficiency for the locus expectedly could not sustain flight. Two of these mutant stocks were carried as homozygotes for 6 mo before a surprising event occurred. aGpdh-I ~° "null" homozygotes were observed in the cultures of flies which flew for extended periods of time like their wild ancestors. Biochemical measurements confirmed that they still lacked any detectable ctGPDH activity. Hence, a selective pressure had produced parallel adaptations in favor of flying ability by a mechanism independent of aGPDH activity. Finally, it is well known that dipteran flies lose the capacity to fly with age (11). This deficiency is accompanied by an ultrastructural deformation of the large flight muscle mitochondria or sarcosomes (12, 13) in addition to age-dependent decreases in state 3 rates of oxidative phosphorylation and respiratory control ratios (RCR) (14). Hence, it was of some interest to examine the aGpdh-I mutants for a number of parameters other than aGPDH in an adult life profile. Because of the importance of the system as a principal sources of energy, and because of the nature of muscle metabolism and atrophy during aging, further investigation of the ultrastructural and biochemical consequences of aGpdh-1 mutations seemed warranted. The observation of genetic adaptation of the mutants by alternative mechanisms for performing the three functions of the aGP cycle suggested the search for possible compensatory mechanisms in the repertoire of metabolic activity. We report here a series of experiments designed to characterize more fully the phenotypic and/or metabolic effects of aGpdh-1 mutations and to search for the metabolic basis of adaptation of the a-Gpdh-I "null" mutants. MATERIALS AND METHODS