Report

Changes in Temperature Preferences and Energy Homeostasis in Dystroglycan Mutants

See allHide authors and affiliations

Science  27 Mar 2009:
Vol. 323, Issue 5922, pp. 1740-1743
DOI: 10.1126/science.1165712

Abstract

Temperature affects the physiology, behavior, and evolution of organisms. We conducted mutagenesis and screens for mutants with altered temperature preference in Drosophila melanogaster and identified a cryophilic (cold-seeking) mutant, named atsugari (atu). Reduced expression of the Drosophila ortholog of dystroglycan (DmDG) induced tolerance to cold as well as preference for the low temperature. A sustained increase in mitochondrial oxidative metabolism caused by the reduced expression of DmDG accounted for the cryophilic phenotype of the atu mutant. Although most ectothermic animals do not use metabolically produced heat to regulate body temperature, our results indicate that their thermoregulatory behavior is closely linked to rates of mitochondrial oxidative metabolism and that a mutation in a single gene can induce a sustained change in energy homeostasis and the thermal responses.

Earth has experienced cooling and warming cycles, and organisms exposed to these climate changes either were exterminated or adapted to survive (1, 2). Animals have thermoregulatory systems to adapt their physiological functions, such as energy utilization, growth, reproduction, and locomotion, in response to the wide range of changes in ambient temperature (35). Although mobile animals commonly select a preferred temperature, the biochemical and metabolic processes that underlie the temperature preference remain poorly understood (57).

We isolated several mutants with aberrant temperature preferences; these included warm-seeking mutants, temperature-insensitive mutants, and the cryophilic mutant, designated as atsugari (atu), described here. On a linear thermal gradient ranging from 12° to 35°C, the third-instar larvae of wild-type Drosophila (Canton S) that had grown at 25°C showed a strong temperature preference that peaked at 22°C (Fig. 1, A and C). The atu mutant larvae had a preference peak at 18°C (Fig. 1, B and D). The behavioral traits of the atu mutant, including assays of olfactory, visual, and locomotory functions, were normal (fig. S1). To exclude the potential effects of the genetic background on the atu mutation, we outcrossed the atu mutant with the isogenic line w1118 and generated P-element excision strains. The atu mutant larvae again exhibited low-temperature preference after outcrossing, and a revertant line with precise P-element excision had a normal temperature preference that peaked at 22°C (Fig. 1E).

Fig. 1.

Identification and characterization of the atu mutant. Representative results of temperature-preference assays. (A) Wild type; and (B) the atu mutant. Arrowheads indicate the region, at 28°C, where the larvae were placed initially. The summed histograms of the larval distributions after 20-min wandering for the wild type (Canton S) (C), the atu mutant (D), revertant line R1 (E), EP2241 line (F), and atu/EP2241 heteroallelic mutant (G). The dotted lines in (D) to (G) represent the Gaussian distribution of the wild type. The numerical analyses of data are shown in table S1. (H) Northern blot of total RNA extracted from third-instar larvae after probing with DmDG cDNA. (I) Expression of DmDG in the atu mutant and the EP2241 line. Western blots of total homogenates prepared from third-instar larvae after incubation with antibodies to DmDG (anti-DmDG) or anti-tubulin are shown.

We cloned the genomic DNA that flanked the P element in the atu mutant. A P element had been inserted 251 base pairs (bp) downstream of the transcription initiation site in the first exon of the Drosophila ortholog of the mammalian gene for dystroglycan (DmDG) (8) (fig. S2, A and B). The inserted P element reduced the expression of the DmDG transcript to 15% of that in wild-type larvae (Fig. 1H). The reduced expression of DmDG in the atu mutant was confirmed with polyclonal antibodies to DmDG (Fig. 1I). Immunohistological analysis showed that DmDG was expressed predominantly in glial cells of the peripheral and central nervous system, epidermis, digestive tract, neuromuscular junction, and ring gland, and the expression of DmDG was severely reduced in all of these tissues of the atu mutant (fig. S3). In the EP2241 mutant fly line that harbors the P-element insertion 330 bp downstream of the transcription initiation site of the DmDG gene (9), the DmDG expression was reduced in a manner similar to that in the atu mutant (Fig. 1I), and the EP2241 mutant exhibited a significant low-temperature preference (Fig. 1F). The heteroallelic combination of atu and EP2241 also induced a low-temperature preference (Fig. 1G). These results indicate that the atu mutation is a hypomorphic allele of the DmDG gene and that the reduced expression of DmDG results in the cryophilic phenotype of the atu mutant.

The preferred temperature of the DmDG transgenic atu line (actin5C-DmDG transgenic atu line), in which the expression of DmDG was driven by the actin5C promotor (fig. S4) in the atu background, was significantly higher than that of the control atu mutant carrying either UAS-DmDG or actin5C-GAL4 (Table 1). To suppress the expression of DmDG in wild-type larvae, we used double-stranded RNA (dsRNA) interference. We constructed a transgenic line carrying UAS-dsRNA and crossed it with the actin5C-GAL4 driver to induce ubiquitous expression of dsRNA specific for DmDG. Immunoblotting with DmDG-specific antibodies showed that the expression of DmDG was reduced to an amount similar to that in the atu mutant (fig. S6). The preferred temperature of the “DmDG-depleted” line was significantly lower than those of the control lines (Table 1). These results confirm that the reduced expression of DmDG induced low-temperature preference in Drosophila.

Table 1.

Reversal of the atu cryophilic phenotype by the transgenic expression of DmDG and pheno-copying by an RNA interference–mediated suppression of DmDG in the wild-type larvae. Comparisons among multiple groups were evaluated by two-way analyses of variance (ANOVAs) followed by the Tukey-Kramer post hoc tests. In each group, values not sharing the same superscript letter are significantly different (P < 0.05). Effects of the cell-specific transgenic expression in the neurons of the anten-nomaxillary complex (19) (figs. S4 and S5) were also examined. The numerical analyses of data are shown in table S1. μ and σ2 denote population mean and population variance, respectively.

View this table:

The tolerance of the atu mutant larvae to cold was greater than that of the wild type (Fig. 2A), though susceptibility to high temperatures was similar in both strains (Fig. 2B). The cold-tolerant phenotype was completely reversed by the transgenic expression of the DmDG transcript (Fig. 2C) and was reproduced in the “DmDG-depleted” line (Fig. 2D). Thus, reduced expression of DmDG causes cold-tolerance as well as preference for low temperatures. Many ectothermic animals compensate for thermal changes by altering their metabolic rate, and the acclimation to low temperature is frequently associated with an increased metabolic rate (3). Indeed, the metabolic rate of the atu mutant, measured in terms of the production of CO2, was almost twice (1.9 times) as high as that of the wild type (Fig. 2E). Concentrations of adenosine 5′-triphosphate (ATP) were significantly increased in the atu mutant (Fig. 2F). The activity of pyruvate dehydrogenase (PDH), a rate-limiting enzyme of the tricarboxylic acid cycle that catalyzes the conversion of pyruvate to acetyl–coenzyme A, was also increased in the atu mutant (2.0 times as high as that of the wild type) (Fig. 2G). The increased metabolic rate, concentrations of ATP, and enzymatic activity of PDH returned toward wild-type levels upon the transgenic expression of DmDG (Fig. 2, E to G), indicating that the reduced expression of DmDG was responsible for the increased energy metabolism. These results raise the possibility that the sustained increase in energy metabolism caused by the reduced expression of DmDG affected the thermoregulatory behavior of Drosophila.

Fig. 2.

Cold tolerance and increased energy metabolism induced by the decreased expression of DmDG. Tolerance of the wild type and the atu mutant to low (A) or high (B) temperatures. Results represent the average of three independent experiments and error bars indicate the SEM. (C to I) The statistical significance of differences between each group was assessed by ANOVA followed by Fisher's protected least significant difference post hoc test (*P < 0.05; **P < 0.01; ***P < 0.001). (C) Cold tolerance of the wild type, the atu mutant, and the atu mutant with transgenic expression of DmDG. The numerical data are shown in (20). (D) Cold tolerance of “DmDG-knockdown” larvae. The numerical data are shown in (21). (E) Metabolic rates as measured by CO2 production. The numerical data are shown in (22). (F) ATP concentrations of the third-instar larvae. The numerical data are shown in (23). (G) PDH activities assessed by NADH (nicotinamide adenine dinucleotide, reduced) production in the mitochondrial fraction. The numerical data are shown in (24). (H) The basal [Ca2+]i in fat body cells. The numerical data are shown in (25). (I) The basal [Ca2+]i in the epithelial cells of the midgut. The numerical data are shown in (26).

To examine whether the increased energy metabolism contributed to the cryophilic phenotype of the atu mutant, we genetically manipulated the activity of PDH using a mutant line with reduced expression of pyruvate dehydrogenase phosphatase (PDP), an activator of PDH (10). The PDP mutant line exhibited a preference for higher temperature than did control lines, and the mutation of the PDP gene completely reversed the atu cryophilic phenotype (Table 2). The cryophilic phenotype of the atu mutant was alleviated to some extent by administering rotenone, a pharmacological inhibitor of oxidative phosphorylation, but not tolbutamide, a sulphonylurea that stimulates metabolic rate by raising the lymphatic concentration of glucose (11) (Table 2). These results indicated that the increased energy metabolism was responsible for the low-temperature preference of the atu mutant. Furthermore, brief exposure of the atu mutant to hyperoxic conditions for 45 min under humid conditions increased preferred temperature of the atu mutant, and the cryophilic phenotype was no longer discernible after the hyperoxic (100% O2) treatment (Table 2). The metabolic rates of the atu mutant were significantly decreased at low temperatures (1.5 ± 0.05 μl CO2 mg–1 hour–1, n = 5 at 18°C; 1.2 ± 0.13 μl CO2 mg–1 hour–1, n = 3 at 14°C), becoming comparable to those observed in the wild type (1.4 ± 0.09 μl CO2 mg–1 hour–1, n = 5 at 18°C; 1.2 ± 0.10 μl CO2 mg–1 hour–1, n = 3 at 14°C). Together, these results indicate that the cryophilic phenotype of the atu mutant is a homeostatic response that depresses the increased oxidative metabolism to establish a steady state in which the increased consumption of oxygen and the supply of oxygen are balanced. Hypoxia elicits a regulated decrease in body temperature in organisms from protozoans to mammals (12). This behavioral thermoregulation is believed to be an adaptation to hypoxia because it lowers the metabolic rate when the supply of O2 is limited, thereby facilitating survival. It seems unlikely that the cryophilic phenotype of the atu mutant is caused by hypoxia, because the expression of the gene for lactate dehydrogenase (LDH), a hypoxia-inducible gene in both Drosophila and mammals, was not increased in the atu mutant (fig. S7). Also, the metabolic rate and the concentrations of ATP generally decrease during hypoxia, whereas both increased in the atu mutant (Fig. 2, E and F).

Table 2.

Reversal of the cryophilic phenotype of the atu mutant by the PDP mutation, a pharmacological inhibitor of oxidative phosphorylation, and hyperoxic treatment. Comparisons among multiple groups were evaluated by ANOVAs followed by the Tukey-Kramer post hoc tests. In each group, values not sharing the same superscript letter are significantly different (P < 0.05). The numerical analyses of data are shown in table S1.

View this table:

An increase in the intracellular Ca2+ concentration ([Ca2+]i) in response to a decrease in the environmental temperature has been widely observed in mammals, ectotherms, and plants, suggesting that [Ca2+]i is a key sign of low temperature (13, 14). An increase in [Ca2+]i, which causes a subsequent uptake of Ca2+ by mitochondria, activates mitochondrial matrix dehydrogenases including PDH, resulting in the acceleration of mitochondrial ATP synthesis (15, 16). In human patients suffering from inherited myopathies, disruption of the dystrophin-dystroglycan complex is suggested to impair the stability of the plasma membrane, resulting in a greater fragility toward mechanical stress and increased permeability to Ca2+, which may lead to the degeneration of muscle fibers (16, 17). To test whether the decreased expression of DmDG might cause an increase in [Ca2+]i, leading to the activation of PDH and the acceleration of the mitochondrial oxidative metabolism, we examined whether [Ca2+]i was affected in the atu mutant. The basal [Ca2+]i of fat body cells from the atu mutant was significantly higher than that of the wild type (Fig. 2H). The increased basal [Ca2+]i in the atu mutant returned to the wild-type value upon the transgenic expression of DmDG, indicating that the reduced expression of DmDG causes a sustained increase in the basal [Ca2+]i in fat body cells. In midgut epithelial cells, the basal [Ca2+]i of the atu mutant was also higher than that of the wild type (Fig. 2I), and the increased basal [Ca2+]i in the epithelial cells of the atu mutant was significantly decreased upon the transgenic expression of DmDG (Fig. 2I). Among mitochondrial dehydrogenases in insects, PDH is the only enzyme that is reported to show Ca2+-dependent activation (18), and therefore, it is possible that the reduced expression of DmDG causes the sustained increase in the [Ca2+]i, which, in turn, induces the activation of PDH, resulting in increased mitochondrial oxidative metabolism. Based on these observations, we propose that a reduction in the DmDG content initiates a chain of sequential reactions, i.e., increased membrane fluidity, activation of Ca2+ influx, elevated mitochondrial metabolism, and eventually, altered thermoregulatory behavior (fig. S9). It remains an enigma, however, how the metabolic changes in a cell are translated into neural code that induces behavioral change at the level of the whole animal (see also supporting online text).

Supporting Online Material

www.sciencemag.org/cgi/content/full/323/5922/1740/DC1

Materials and Methods

SOM Text

Figs. S1 to S11

Tables S1

References

References and Notes

View Abstract

Stay Connected to Science

Navigate This Article