Distinguishing PEX gene variant severity for mild, severe, and atypical peroxisome biogenesis disorders in Drosophila

Abstract

Peroxisomal biogenesis disorders (PBD) are autosomal recessive disorders caused by loss-of-function mutations of one of the PEX genes responsible for peroxisomal formation. Impaired peroxisome assembly causes severe multisystemic failure with patient phenotypes ranging from epilepsy, liver disease, feeding issues, biochemical abnormalities, and neurodegeneration. Variants in the same PEX gene can produce wide differences in severity, ranging from individuals with death in the first year of life to adults with milder complications. To study this strong genotype-phenotype correlation, we selected specific human PEX gene mutations and utilized Drosophila as a model organism. We generated flies replacing the coding sequence of our Pex gene of interest with a KozakGAL4 (KZ) promoter trap sequence. These cassettes simultaneously knock-out of the Pex gene and knock-in a GAL4 driver, ideal for making "humanized" flies in which the human PEX gene can replace the fly loss. We assessed Pex2 ^KZ and Pex16 ^KZ lines in lifespan, bang sensitivity, and climbing assays and confirmed that these are strong loss-of-function alleles. In parallel, we generated human reference and variant UAS-cDNA lines of PEX2 and PEX16 variants in Drosophila. We observed nearly complete phenotypic rescue of Drosophila Pex2 and Pex16 loss when human PEX2 ^Ref or PEX16 ^Ref , respectively, were expressed. We also provide evidence for an allele severity spectrum in PEX2 and PEX16 in which some missense alleles, such as PEX2 ^C247R , are equally severe as early truncations, such as PEX2 ^R119*. We also observed that alleles associated with mild PBD, such as PEX2 ^E55K , show variability depending on the assay but do not fully rescue. Finally, alleles associated with atypical ataxia phenotypes, such as PEX16 ^F332Del , can perform as well as PEX16 ^Ref , depending on the assay. Altogether, these Drosophila lines effectively model the range of severity of peroxisomal biogenesis disorders.

Figure 1

The KozakGAL4 knock-in/knock-out strategy, which replaces the coding region of the Pex gene of interest by identifying sgRNA target sites in the 5’ UTR and 3’ UTR. Gray boxes, UTRs; blue boxes, Pex-coding region. (A) Schematics of Pex2 locus and targeting strategy. (B) Schematics of Pex16 locus and targeting strategy. (C) Expression pattern of Pex2^KZ in 3rd instar larval brain. (C’) Repo expression. (C”) Elav expression. (C”’) Merged image with co-localization within the ventral nerve cord, indicating Pex2^KZ expression in both neurons (Elav) and nuclear glia (Elav) in the larval brain. (D) Expression of Pex16^KZ expression pattern in the third instar Drosophila larval brain. (D’) Repo expression. (D”) Elav expression. (D”’) Merged image with co-localization within the ventral nerve cord, indicating Pex16^KZ expression in both neurons (Elav) and nuclear glia (Elav) in the larval brain.

Figure 2

Drosophila Pex2 mutants have shortened lifespan, are bang-sensitive, and have a climbing defect. (A) Schematic representation of fly Pex2 gene along with four alleles, including a coding P-element insertion (Pex2^EPg), 2 deletion alleles (Pex2¹, Pex2²), and Pex2-KozakGAL4 (Pex2^KZ). (B) Pex2 female lifespan assay shows that the Pex2 mutants have a shorter lifespan compared to control lines (pink and purple). (C) Pex2 male lifespan assay shows that the Pex2 mutants have a shorter lifespan compared to control lines, worse than female mutant flies. (D) Pex2 null flies have a significant bang-sensitive phenotype (red) compared to controls (pink and purple) observed at 10 days after eclosion (DAE). (E) Pex2 null flies have a significant climbing deficiency (red) compared to controls (pink and purple) observed at 10 days after eclosion. [* = p-value is less than 0.05. ** = p-value is less than 0.01. *** = p-value is less than 0.001. **** = p-value is less than 0.0001]

Figure 3

Human UAS cDNA PEX2 reference and variant lines. (A) Schematic representation of human PEX2 gene. (B) Schematic representation of human PEX2 protein and variant locations. (C) PEX2 variant table indicates the consequence of the change, pathogenicity prediction, clinical significance, clinical severity in homozygosity and heterozygosity, and conservation in Drosophila. (D) Indicates the observed/expected Mendelian ratio of the F1 generation of human PEX2 variants, in a fly null background. (E) Assessment of the phenotype of indicated genotypes as lethal, viable, or semi-lethal.

Figure 4

Human PEX2 expression in Pex2 null background larvae significantly rescues peroxisomes in 3^rd instar larva body wall 6 muscle. (A, A’& A”) represents control group; (B, B’, & B”) represents Pex2 null flies & (C, C’ & C”) represents human rescue group. Quantification of Pex3 positive puncta between the three genotypes. [*p<0.05; ****p<0.0001]

Figure 5

Rescue-based humanization of Pex2: Behavior Assays. (A) Lifespan analysis of PEX2^{Ref and Variant} female flies, along with our Pex2 null and control lines. (B) Lifespan analysis of PEX2^{Ref and Variant} male flies, along with our Pex2 null and control lines. (C) Bang sensitivity assay of PEX2^{Ref and Variant} female flies, along with our Pex2 null and control lines, at 10 days after eclosion (DAE). (D) Bang sensitivity assay of PEX2^{Ref and Variant} female flies, along with our Pex2 null and control lines, at 15 days after eclosion. (E-G) Climbing assay of PEX2^{Ref and Variant} female flies, along with our Pex2 null and control lines, at 5 days, 10 days, and 15 days after eclosion. [* = p-value is less than 0.05. ** = p-value is less than 0.01. *** = p-value is less than 0.001. **** = p-value is less than 0.0001]

Figure 6

Human PEX2 expression in Pex2 null background significantly rescues peroxisome number in direct flight muscle (DFM49) in adult flies. (A, A’, A” & A”’) represents control group; (B, B’, B” & B”’) represents Pex2 null flies & (C, C’, C” & C”’) represents human rescue group. (A”“, B”’ & C”“) illustrates the closure look at Pex3 puncta within the square box selected ROI for each respective group i.e., control, knockdown & rescue. Scale bar corresponds to 10 μm & 5 μm respectively. DFM49 marked with long dashed line. (D) Quantification of Pex3 puncta between the three genotypes. [**p<0.01; ****p<0.0001]