Mediators of a long-term movement abnormality in a Drosophila melanogaster model of classic galactosemia

Abstract

Despite neonatal diagnosis and life-long dietary restriction of galactose, many patients with classic galactosemia grow to experience significant long-term complications. Among the more common are speech, cognitive, behavioral, ovarian and neurological/movement difficulties. Despite decades of research, the pathophysiology of these long-term complications remains obscure, hindering prognosis and attempts at improved intervention. As a first step to overcome this roadblock we have begun to explore long-term outcomes in our previously reported GALT-null Drosophila melanogaster model of classic galactosemia. Here we describe the first of these studies. Using a countercurrent device, a simple climbing assay, and a startle response test to characterize and quantify an apparent movement abnormality, we explored the impact of cryptic GALT expression on phenotype, tested the role of sublethal galactose exposure and galactose-1-phosphate (gal-1P) accumulation, tested the impact of age, and searched for potential anatomical defects in brain and muscle. We found that about 2.5% residual GALT activity was sufficient to reduce outcome severity. Surprisingly, sublethal galactose exposure and gal-1P accumulation during development showed no effect on the adult phenotype. Finally, despite the apparent neurological or neuromuscular nature of the complication we found no clear morphological differences between mutants and controls in brain or muscle, suggesting that the defect is subtle and/or is physiologic rather than structural. Combined, our results confirm that, like human patients, GALT-null Drosophila experience significant long-term complications that occur independently of galactose exposure, and serve as a proof of principle demonstrating utility of the GALT-null Drosophila model as a tool for exploring genetic and environmental modifiers of long-term outcome in GALT deficiency.

Fig. 1.

Movement abnormality in GALT-null D. melanogaster. The proportion of flies from each indicated cohort to reach the tenth (final) chamber of a countercurrent device. All values were modeled in a one-way ANOVA with a variable to control for day-to-day variations in testing. dGALT^KG00049 carries a P-element insertion just upstream of the Drosophila GALT gene that does not interfere with GALT expression or activity. dGALT^ΔAP2 is a nearly full deletion of the Drosophila GALT gene achieved by imprecise excision of the P-element from dGALT^KG00049 (Kushner et al., 2010). dGALT^C2 is a wild-type allele generated by precise excision of the same P-element (Kushner et al., 2010). In flies with endogenous GALT enzyme activity (dGALT^C2 and dGALT^KG00049) more than 60% of each cohort reached the final chamber. Flies lacking GALT activity (dGALT^ΔAP2 and dGALT^ΔAP2 with Actin5c-GAL4) had difficulty with this behavior and on average only about 20% reached the final chamber. Expression of a human GALT transgene rescued the phenotype. Significant differences are indicated. All values represent mean ± s.e.m. n, number of cohorts tested.

Fig. 2.

GALT-null D. melanogaster are defective in climbing but not startle response. (A) The proportion of flies in each indicated cohort that reached above a designated mark in 20 seconds. (B) The average number of seconds it took all of the flies in each cohort to regain a standing posture after vigorous agitation for 10 seconds. All values were modeled in a one-way ANOVA and mean ± s.e.m. are plotted. n, number of cohorts tested.

Fig. 3.

Relationship between GALT activity and a movement abnormality in D. melanogaster. The ability of animals of each indicated genotype to reach the tenth chamber of a countercurrent device is plotted as a function of their adult GALT enzymatic activity. As GALT activity increased from null toward wild-type levels, an animal’s ability to reach the tenth chamber improved. There was no apparent impact from GALT overexpression. All values were modeled in a one-way ANOVA with a variable to control for day-to-day variations in testing. All values represent mean ± s.e.m. n, number of cohorts tested.

Fig. 4.

Low-level galactose exposure during development has no impact on a movement abnormality in adult GALT-null D. melanogaster. (A) GALT-null and control animals were tested in the countercurrent device after being raised on a diet of food containing 555 mM glucose as the sole monosaccharide (white bars), or 555 mM glucose supplemented with 50 mM galactose (shaded bars). (B) Gal-1P accumulation in GALT-null and control larvae, newly eclosed adults or adults transferred following eclosion to glucose-only food for 2 days prior to analysis. As described in A, some animals were exposed during development to food containing glucose as the sole monosaccharide (white bars), whereas others were exposed to food containing glucose spiked with 50 mM galactose (shaded bars). Late-stage larvae and adult animals were collected and analyzed as described in Methods. Gal-1P levels were standardized to protein concentration. All values represent mean ± s.e.m. n, number of cohorts tested.

Fig. 5.

Impact of age on movement in adult GALT-null D. melanogaster. GALT-null (shaded bars) and control flies (white bars) were tested using the countercurrent device at ages 2, 9 and 16 days post-eclosion. Whereas the control flies demonstrated a gradual, progressive decline with age the GALT-null flies demonstrated a decline that was notably accelerated and profound. All values represent mean ± s.e.m. n, number of cohorts tested.

Fig. 6.

GALT-null D. melanogaster show no apparent morphological defects in brain or muscle. (A) Representative images of adult brain (upper panels) and muscle tissue (lower panels) from GALT-null and control animals. Arrows in the brain images point to cortex, arrowheads point to vacuoles and the asterisk indicates normal neuropil. There were no apparent differences in gross anatomical structures between the GALT-null and control animals in either tissue. Scale bar: 20 μm. (B) Representative immunofluorescence stains for mitochondria using an antibody directed to ATP synthase reveals no clear abnormalities of mitochondrial structure or number in GALT-null neurons. A DAPI counterstain was used to stain nuclei. Scale bar: 5 μm. (C) Representative immunohistochemical stains for synapsin (SYN), the vesicular glutamate transporter (VGLUT) and the microtubule binding protein futsch revealed no clear defects in GALT-null animals. Scale bar: 40 μm.