Transcriptome and functional analysis in a Drosophila model of NGLY1 deficiency provides insight into therapeutic approaches

Abstract

Autosomal recessive loss-of-function mutations in N-glycanase 1 (NGLY1) cause NGLY1 deficiency, the only known human disease of deglycosylation. Patients present with developmental delay, movement disorder, seizures, liver dysfunction and alacrima. NGLY1 is a conserved cytoplasmic component of the Endoplasmic Reticulum Associated Degradation (ERAD) pathway. ERAD clears misfolded proteins from the ER lumen. However, it is unclear how loss of NGLY1 function impacts ERAD and other cellular processes and results in the constellation of problems associated with NGLY1 deficiency. To understand how loss of NGLY1 contributes to disease, we developed a Drosophila model of NGLY1 deficiency. Loss of NGLY1 function resulted in developmental delay and lethality. We used RNAseq to determine which processes are misregulated in the absence of NGLY1. Transcriptome analysis showed no evidence of ER stress upon NGLY1 knockdown. However, loss of NGLY1 resulted in a strong signature of NRF1 dysfunction among downregulated genes, as evidenced by an enrichment of genes encoding proteasome components and proteins involved in oxidation-reduction. A number of transcriptome changes also suggested potential therapeutic interventions, including dysregulation of GlcNAc synthesis and upregulation of the heat shock response. We show that increasing the function of both pathways rescues lethality. Together, transcriptome analysis in a Drosophila model of NGLY1 deficiency provides insight into potential therapeutic approaches.

Figure 1.

Ubiquitous knockdown of dNGLY1 results in lethality. (A) Proportion of expected dNGLY1 knockdown flies collected at adult eclosion. Replicates for two different RNAi lines are shown. (B) Developmental delay is observed in dNGLY1 knockdown flies compared with control flies. Representative results of more than four experiments are shown for both RNAi strains and their respective controls. RNAi expression is driven by the ubiquitous tubulin-GAL4 driver.

Figure 2.

Numerous genes are misregulated in dNGLY1 knockdown flies. Standard volcano plot showing the genes that are misregulated in dNGLY1 knockdown flies as compared with control flies, as measured by RNAseq. Relevant genes are labeled. Red=genes with 1.5-fold change and at an FDR of 5%. q value=FDR.

Figure 3.

GlcNAc supplementation partially rescues dNGLY1 knockdown lethality. (A) Expression changes in components of the UDP-GlcNAc biosynthetic pathway between dNGLY1 knockdown and control flies. Gfat1 is the only component that shows a significant change. Gfat1 is nearly 2-fold downregulated in dNGLY1 knockdown flies (q=1.7×10⁻¹⁰). Human orthologs of Drosophila components shown above. Black horizontal line represents 1.5-fold downregulation. (B) Dietary GlcNAc supplementation during development rescues developmental delay (P<0.002). Cumulative number of flies counted on each collection day is plotted. (C) At the end of 11 days of collecting, there was a significant increase in the proportion of expected dNGLY1 knockdown flies raised on GlcNAc, when compared with no GlcNAc supplementation (P<2.4×10⁻²³). All data are representative of at least three independent experiments.

Figure 4.

GlcNAc supplementation improves longevity in dNGLY1 knockdown adult flies. (A) GlcNAc supplementation scheme used during larval development or adult maintenance (0 or 100 μg/ml). Colors represent the different combinations. Adult GlcNAc supplementation improves longevity, irrespective of larval supplementation status. Comparisons of dNGLY1 knockdown adult longevity (days post-eclosion) for (B) no GlcNAc supplementation versus complete GlcNAc supplementation (P=0.021); (C) no GlcNAc supplementation versus adult only supplementation (P=0.013); (D) larval only GlcNAc supplementation versus complete supplementation (P=0.049). GlcNAc supplementation had no significant effect on control fly longevity (B–D; right).

Figure 5.

Heat shock rescues lethality in dNGLY1 knockdown flies. (A) Upregulation of heat shock genes in dNGLY1 knockdown flies when compared with control flies. All genes shown are significantly upregulated (q<0.05). Black horizontal line=1.5-fold upregulated. (B) Heat shock treatment during larval development recues lethality. Heat shock during each of the three larval instar stages significantly improves survival (first instar: P=6.9×10⁻⁶; second instar: P=1.7×10⁻⁹; third instar: P=2.4×10⁻⁵). None=no heat shock treatment. (C) Heat shock genes are required for survival in dNGLY1 knockdown flies. Knockdown of Hsp70Bb (P<3.3×10⁻¹²), Hsp23 (P<8.5×10⁻⁵) and Hsp26 (P<4.8×10⁻⁵) on the dNGLY1 knockdown background significantly reduced survival. Knockdown of Hsp68 on the dNGLY1 knockdown background did not impact survival. These experiments were performed under normal developmental conditions, without heat shock treatment. (D) Ubiquitous knockdown of Hsp70, Hsp23, Hsp26 or Hsp68 alone had no effect on survival. Knockdown of Hsp83 alone resulted in near complete lethality (P=1.7×10⁻¹⁸). These experiments were performed under normal developmental conditions, without heat shock treatment. Mean±standard deviation. All data represent at least three independent experiments.