(A) Northern blot of whole flies using full-length pink1 as a probe. Compared with wildtype, pink1 null mutant (pink1⁵) flies do not show full length mRNA. Transgenic flies expressing pink1 G426D mutants show comparable or higher expression of pink1 than is seen in flies expressing wildtype pink1. Arrow points to pink1 expression and rp49 (*) serves as an RNA loading control.
(B–J) Schematic and phase contrast micrographs of spermatid mitochondria during the "onion stage" (C–F) and "the leaf blade stage" (G–J). Compared with wildtype, pink1 mutants show vacuolation (red arrowhead) in the nebenkern during the onion stage (D), and one instead of two mitochondrial derivatives (yellow arrow) seen in wildtype at the leaf blade stage (H). A genomic rescue transgene carrying wildtype pink1 (CaSpeR-pink1) completely rescues the male sterility due to lack of pink1 (100% fertile, n>120) (Clark et al., 2006), and spermatid phenotypes in both the onion stage and the leaf-blade stage (F,J). In contrast, pink1G426D (CaSpeR-pink1G426D) fails to rescue the sterility of pink1 males (<2% fertile, n=60), or the spermatid phenotypes (E,I). Red arrowheads point to vacuolation of the nebenkern and orange arrows mark each mitochondrial derivative.
(K–N) Mitochondria of indirect flight muscle are labeled by mito-GFP. Compared with control (K), pink1 mutants display overall reduced levels of mitoGFP signal, and large clumps of intense GFP signal (L), which can be completely rescued by overexpression of Pink1WT (N), but not Pink1G426D (M).

(A–C) Anti-Tyrosine Hydroxylase immunostaining of whole-mount brains from 40-day old wildtype (A), pink1 mutant (B) and omi mutant flies (C). 17–27 individual flies (both hemispheres) were counted for each genotype. (D) Schematic depicting locations of the major dopaminergic neuron clusters in the adult brain as designated by abbreviations (Nassel and Elekes, 1992). The major dopaminergic neuron clusters are located near the posterior surface of the brain, with the exception of the PAL, which is located near the anterior surface (labeled in gray). (E,F) Quantification of dopaminergic neurons in each cluster in pink1 mutant flies and wildtype flies (E), omi mutant flies and wildtype flies (F) aged for 40 days at 25°C. Error bars represent standard deviations, and student T-test is used for statistical analysis. Toluidine Blue staining of indirect flight muscle fibers (G,I,K) and EM studies of these muscles (H,J,L). The borders of mitochondria are marked with white dashed lines. In contrast to pink1 mutants (I,J), omi mutants do not show muscle degeneration or mitochondrial morphological defects (K,L), even when aged for 30 days. Genotypes: wildtype: w/Y. omi mutant: w/Y; omi^NSO /Df(3R)ED5644. pink1 mutant: w pink1⁵/Y. Scale bars: 1µm in F,H,J.

(A–F) Double labeling of onion-staged spermatids in wildtype (A–C) and pink1 mutants (D–F) using Prosα6T-GFP (green), which labels the nucleus, and an anti-Omi antibody (red), which labels the Nebenkern. In pink1 null mutants, Omi is still localized to the nebenkerns of spermatids. (G) Western blotting of endogenous Pink1-9Myc expression using a genomic rescue transgene. Loss of omi function does not alter the cleavage pattern of Pink1.

Scanning EM and light micrographs of Drosophila eyes. At 18°C, omi overexpression (B) results in wildtype-appearing eyes, whereas overexpression of pink1 (C) leads to mild rough eyes. Overexpression of both omi and pink1 results in a small and rough eye (F). This pink1-omi overexpression interaction is abolished with expression of a protease inactive version of Omi, S266A (G). Expression of OmiS266A by itself does not result in any eye phenotypes (D). Expression of a mutation analogous to the Pink1 disease mutant, G426D, has no phenotype (E), however, it still shows an interaction with omi overexpression (H). The phenotype of pink1 overexpression cannot be suppressed by loss of omi function induced by RNAi-omi, even when raised at 29°C (I, J). Silencing of omi function by RNAi-omi is strong since it completely suppresses the omi overexpression induced eye phenotype (Supplementary Fig. S2). Similarly, the eye phenotype due to omi overexpression cannot be modified by lack of pink1 (K,L). Panels A–H are from flies raised at 18°C, whereas panels I–L are from flies raised at 29°C. Genotypes: Control: w; GMR-Gal4/+. Omi: w; GMR-Gal4, UAS-omi/+. Pink1: w; GMR-Gal4, UAS-pink1/+. Pink1G426D: w; GMR-Gal4/UAS-Pink1G426D. Omi+Pink1: w; UAS-omi/+; GMR-Gal4, UAS-pink1/+. Omi+Pink1G426D: w; GMR-Gal4, UAS-omi/UAS-Pink1G426D. Pink1+OmiS266A: w; UAS-OmiS266A/+; GMR-Gal4, UAS-pink1/+. Pink1+omi lof: w; GMR-Gal4, UAS-pink1/UAS-RNAi-omi. Omi+pink1 lof: w pink1⁵; GMR-Gal4, UAS-omi/+.

(A) Schematic depicting domains of Omi. MTS: mitochondrial targeting sequence. TM: transmembrane domain. IBM: IAP binding motif. The exact locations of omi^NSO and omi^V110E are depicted as a blue and a red asterisk, respectively. (B) Sequence alignment of Omi/HtrA2 in various species in a highly conserved region, in which the conserved Valine, mutated in omi^V110E, is marked in red. (C–E, C’–E’) Phase contrast micrographs of testes. In contrast to a control fly (C, C’) in which the seminal vesicles (arrow) is full of sperm (phase dark), omi mutants (D, D’) show empty seminal vesicle (bracket with an asterisk), which can be rescued by omi overexpression (E, E’) (arrow pointing to the seminal vesicle). C’–E’ are higher magnification views of the seminal vesicle from C–E. (F–H) Phalloidin staining of investment cones within one syncytial cyst. In contrast to controls in which investment cones are well aligned indicating synchronized movement (F), omi mutants show scattered investment cones in some of the cysts, indicative of a mild defect in individualization (G), which can be rescued by omi overexpression (H). (I–P) Schematics and phase contrast micrographs of spermatid mitochondria during the "onion stage" (I–L) and "the leaf blade stage" (I, M–O). Compared with wildtype (J,M), omi mutants do not show any defects in either stage (K,N), whereas pink1 mutants show vacuolation (red arrowhead) in the nebenkern during the onion stage (L), and one instead of two mitochondrial derivatives (yellow arrow) seen in wildtype at the leaf blade stage (O). (P–S) Schematic and transmission EM images of a portion of a post-individualization cyst. Each spermatid contains an axoneme (orange arrow) and mitochondrial derivative (red arrowhead) within an individual plasma membrane. The omi mutant cyst (R) shows disorganization of spermatids and occasional individualization defects (data not shown). However, compared with pink1 mutants, which show severe impairment in the size and morphology of mitochondria (S), omi mutant cysts show normal appearing mitochondria (R). Genotypes: wildtype: w/Y; control: w/Y; omi^NSO/+; omi mutant: w/Y; omi^NSO/Df(3R)ED5644; omi mutant + rescue: w/Y; TMR-omi/+; omi^NSO/Df(3R)ED5644. Scale bars: 500µm in C–E; 500nm in Q–S.

(A–H) Phase contrast micrographs of spermatid mitochondria during the "onion stage" (A–E) and "the leaf blade stage" (F–J). pink1 null mutants show vacuolation of nebenkerns (B) and one single mitochondrial derivative (G). This phenotype can be completely suppressed by pink1 overexpression (C,H), but not by omi overexpression (D,I). Double mutants removing both pink1 and omi function result in pink1 mutant-like phenotypes without any enhancement (E,J). Red arrowheads point to vacuolation of the nebenkern and orange arrows mark each mitochondrial derivative. (K–N) Mitochondria of indirect flight muscle are labeled by mito-GFP. Compared with control (K), pink1 mutants display overall reduced levels of mitoGFP signal, and large clumps of intense GFP signal (L), which can be completely rescued by pink1 overexpression (M). Double mutants of pink1 and omi show pink1 mutant-like phenotypes (N). Genotypes: (A,F) w. (B,G) w pink1⁵/Y. (C,H) w pink1⁵/Y; TMR-pink1/+. (D,I) w pink1⁵/Y; TMR-omi/+. (E,J) w pink1⁵/Y; omi^NSO/Df(3R)ED5644. (K) FM6/Y; Mef2-Gal4, UAS-mitoGFP/+. (L) w pink1⁵/Y; Mef2-Gal4, UAS-mitoGFP/+. (M) w pink1⁵/Y; Mef2-Gal4, UAS-mitoGFP/UAS-pink1. (N) w pink1⁵/Y; Mef2-Gal4, UAS-mitoGFP/UAS-RNAi-omi.

(A) Schematic depicting positions of 4 mutations (*) with respect to domains in Omi. (B–C) Phase contrast micrographs. As with overexpression of Omi WT (Fig. 3E,E'), expression of G363S (C), but not S266A (B), restores the production of motile sperm in the seminal vesicle of omi mutants. An arrow in C indicates the presence of sperm (phase dark), whereas a bracket and an asterisk in B point to the absence of sperm. (D) Both OmiS236C and OmiS266A are expressed at a comparable levels compared with Omi WT, as detected by an anti-Omi antibody. Western blots of head lysates from flies overexpressing OmiWT, Omi236C or S266A using anti-Omi antibodies. Overexpression is accomplished using the eye-specific driver (GMR-Gal4), and flies are raised at 18°C to avoid cell death compromising recovery of proteins. A non-specific band (*) serves as protein loading control.
(E–J) Scanning EM micrographs of Drosophila eyes. Compared with overexpression of Omi WT, which results in small and rough eyes at 25°C (H), overexpression of Omi G363S or S364A leads to similar rough eye phenotypes (I,J), while overexpression of Omi S236C or S266A results in wildtype-appearing eyes (F,G). GMR-Gal4 is used as an eye-specific driver. Scale bars: 500µm in B,C.