A. Genetic cross to check if Piwi is an enhancer of the eye outgrowth phenotype caused by ectopic expression of Krüppel where CyO or CyO-GFP balancer chromosomes carries a piwi⁺ allele. Reverse crosses with piwi⁻¹ and piwi⁻² males are not shown here. B. Light microscopic images of the adult fly eyes with various types of ectopic outgrowths (black arrows). Images of the eyes of control flies with wild type levels of Piwi are shown in the upper panel. C. Over-expression of maternal Piwi suppresses eye outgrowths when Hsp90 is inhibited. From each cross 50 flies with Kr^If-1/Kr^If-1 background were collected and scored for the phenotype. Experiment was repeated five times with five independent crosses. Average of five independent crosses with standard deviations (s.d.) are plotted. Unpaired t test was performed to calculate statistical significance. * represents a p value of 0.0254 D. Genetic linkage between Piwi and Hsp90 in mediating canalization.

A. Fractionation scheme for identifying peptides interacting with Piwi. B. Fraction #27 obtained from Superdex 200 chromatography was resolved on a 7.5% SDS polyacrylamide gel and stained with silver stain. Identities of individual bands were obtained by excising bands from a colloidal coomassie blue stained gel (not shown here) followed by mass spectrometry. Peptides identified but not relevant to this study are marked by (*). C. Western blotting analysis showing the co-migration of Piwi and Hop in Superdex 200 column. Fraction numbers are marked above and fraction corresponding to ~150kDa is marked below. D. Co-immunoprecipitation of Piwi and Hsp90 with Hop. Control reactions (-IP) did not contain Hop specific antibody. E. Co-immunoprecipitation of Piwi with Hsp90. Control reactions (-IP) did not contain Hsp90 specific antibody. F. Piwi, Hop and Hsp90 function in the same complex. Left panel shows the scheme of the serial immunoprecipitation experiment. Right panel shows western blotting analysis of Piwi, Hop and Hsp90 after the first and second immunoprecipitations. Control reactions (-Myc IP and –HA IP) contained protein A/G plus agarose beads.

A. Genetic cross to test if Hop is a maternal enhancer of the outgrowths. Reverse cross with Hop^k00616 males is not shown here. B. Kr^If-1/Hop^k00616 fly eyes exhibiting the outgrowths (arrows). C. Quantification of the outgrowths observed in Kr^If-1/Hop^k00616 flies. 479 flies were collected from three independent crosses and scored for the outgrowths. Average percentage of flies with the outgrowths and s.d are plotted. The error bar indicates s.d.

A. Genetic crosses for Hop^k00616 selection experiment. Similar cross was setup with piwi⁻¹. Hop^k00616 and piwi⁻¹ mutations are present only in the F1 Kr^If-1 flies. A single male F1 fly with outgrowth was crossed with virgin wild type Canton S female flies to remove Hop^k00616 and piwi⁻¹ mutations. From F4 generation, Kr^If-1/Kr^If-1 males and females containing the outgrowths were selected and inter crossed. B. Quantification of flies with the outgrowths in each generation. 100 flies/generation were scored and counts of male and female flies with the outgrowths are individually plotted. C. Over-expression of wg is ‘fixed’ over multiple generations. Krüppel and wingless mRNA expression in the heads of 5 F8 males and 5 F8 females with eye outgrowths were quantified by qPCR, with the same number of F8 flies without the eye outgrowths as controls. Average values of three independent experiments and s.d. are plotted. Statistical significance was calculated using a paired t test, with p values that are less than 0.05 and 0.001 indicated by * and **, respectively.

A. Hsp90 inactivation by geldanamycin does not change Piwi protein levels. Two-fold serial dilutions of total ovary lysate were resolved by SDS polyacrylamide gel electrophoresis followed by western blotting using various antibodies as shown on the right side of the panel. Non-specific control for the amount of protein loaded in each lane is a signal from a cross reaction of B-Raf antibody to an abundant protein in the ovary lysate. B. Piwi levels do not change in Hsp83 mutants. Two-fold serial dilutions of ovary lysates from either Hsp83⁰⁸⁴⁴⁵/TM3 or Hsp83⁰⁸⁴⁴⁵/Hsp83⁰⁸⁴⁴⁵ flies were resolved by SDS polyacrylamide gel electrophoresis and analyzed by western blot. Notice a decrease in the levels in Hsp90 but not Piwi. As a measure of total protein in each lane, coomassie staining of the most abundant protein (~60 kDa) in the ovary lysate is shown. C. Hsp90-dependent phosphorylation of Piwi. Western blot analysis of second dimension SDS PAGE gel electrophoresis using anti-Piwi antibody is shown. Directions of the first and second dimensions are indicated on the upper left corner of the panel. Presence or absence of geldanamycin and various genotypes of flies used in this study are marked on top of each panel. CIP represents calf intestinal phosphatase. Different isoforms of Piwi are marked from ‘1’ through ‘4’ with ‘1’ being the most positive isoform of Piwi. White arrow head in the second panel represents the most positive Piwi isoform that is enriched in the presence of geldanamycin. Black arrow in the third panel represents the isoform of Piwi that is depleted in Hsp83⁰⁸⁴⁴⁵/Hsp83⁰⁸⁴⁴⁵ mutants. Compare it with dashed black arrow in the fourth panel. Black arrow head in the fourth panel represents an isoform of Piwi that gets enriched in Hsp83⁰⁸⁴⁴⁵/Hsp83⁰⁸⁴⁴⁵ ovaries. Black circle in the third panel marks the area where two isoforms that are depleted upon CIP treatment in the fifth panel (dashed circle) D. Immunoprecipitation of Piwi with anti-phospho-serine, anti-phospho-threonine and anti-phospho-tyrosine antibodies from wild type ovarian lysate. Band representing Piwi is marked on the right. Levels of IgG(H) were used to monitor loading in each lane. Lower panel is a darker exposure of the upper one to show that absence of Piwi in phosphorthreonine IP is not due to less loading E. Immunoprecipitation of Piwi with anti-phospho-serine and anti-phospho-tyrosine antibodies in the presence or absence of CIP treatment. Notice that Piwi is mostly depleted from the immunoprecipitates upon CIP treatment, further confirming that Piwi is phosphorylated. F. Immunoprecipitation of Piwi with anti-phospho-serine and anti-phospho-tyrosine antibodies from wildtype and Hsp83⁰⁸⁴⁴⁵/Hsp83⁰⁸⁴⁴⁵ ovarian lysate. Notice depletion of Piwi from both anti-phospho-serine and anti-phospho-tyrosine IP from Hsp83⁰⁸⁴⁴⁵/Hsp83⁰⁸⁴⁴⁵ mutant ovarian lysate.

See text for details.