Novel role of specific Tudor domains in Tudor-Aubergine protein complex assembly and distribution during Drosophila oogenesis

Abstract

Germ cells give rise to the next generation and contain ribonucleoprotein particles, germ granules. In these granules, Piwi protein Aubergine has been shown to interact with Tudor protein in Drosophila. Tudor protein has 11 Tudor domains and it has been unclear to what extent all these domains are involved in the interaction with Aubergine. Here we present direct biochemical evidence that Tudor-Aubergine interaction surface is composed of different Tudor domains including those that have not been previously implicated in Aubergine recognition. Furthermore, we show that specific single Tudor domains determine localization of Tudor complex to different sites in ovarian germ cells. Our data suggest that multiple Tudor domains of germline proteins from various species are redundantly used for interaction with the same protein partner during germline development.

Fig. 1

Tud mutant proteins analyzed in this study. Wild-type (wt) Tud protein and proteins with mutant Tud domains (empty squares) are shown. tud^A36 and tud^B42 had mutations changing equivalent arginines in Tud domains 1 and 10 respectively (Arg91Trp in tud^A36; Arg2228Cys in tud^B42); tud⁴ had a deletion of three amino acids in Tud domain 7 (Asp1708, Ile1709, Lys1710); tud^B45 lacked Tud domains 10 and 11; tud^A7 had a deletion of a small C-terminal segment caused by premature UAG stop codon; mini-Tud Δ3 lacks Tud domains 2 - 6 [5].

Fig. 2

Involvement of different Tud domains in Aub binding. (A) Mutations in single Tud domains 1, 7, and 10 in full-length (FL) Tud protein do not affect Aub binding in coimmunoprecipitation experiments with anti-Aub antibody. Western blot detections of bound Tud or Aub are shown. Top panel demonstrates levels of Tud and Aub present in extracts from wt and mutant ovaries. tud¹ is a protein-null. Bottom panel shows coimmunoprecipitation of Tud and Aub from wt and Tud domain mutants. As a control, rabbit IgG instead of anti-Aub antibody was added to wt extracts. (B) in vitro direct binding experiments using immunopurified FL Tud protein, mini-Tud Δ3 and functional GFP-Aub. Tud proteins were added to GFP-Aub-associated beads and, as a control, to Protein A beads. Bound Tud and Aub were detected with anti-HA and anti-Aub antibodies respectively. The total (input) amounts of Tud proteins are indicated. From three experiments using equimolar amounts of FL Tud and mini-Tud, binding of mini-Tud to Aub was 2.61% of that of FL Tud to Aub (standard error was 1.69). (C) This mini-Tud Δ3/Aub binding experiment was performed as described for (B) except the input amount of mini-Tud was ~40-fold higher.

Fig. 3

Mutations in single Tud domains of full-length Tud protein differently affect Tud accumulation in ovarian nuage or germ plasm but do not interfere with Aub localization. tud, aub mutants and wild-type (wt) stage 7 egg chambers (Nuage columns) and stage 10 oocyte posterior (Germ plasm columns) were co-stained with rabbit anti-Tud (green) and rat anti-Aub (red) antibody. Overlay images are shown. Anterior is to the left. tud^A36, tud⁴ and tud^B42 mutants had mutations in specific Tud domains (Fig. 1); tud^B45 lacked Tud domains 10 and 11 and tud^A7 had a deletion of a small C-terminal segment (Fig. 1). In the bottom wt row, closer views of wt nuage and germ plasm are shown. Tud is not localized to nuage and oocyte germ plasm in protein-null aub^N11/aub^HN2 ovary (aub^mut).