Squid, Cup, and PABP55B function together to regulate gurken translation in Drosophila

Abstract

During Drosophila melanogaster oogenesis, the proper localization of gurken (grk) mRNA and protein is required for the establishment of the dorsal-ventral axis of the egg and future embryo. Squid (Sqd) is an RNA-binding protein that is required for the correct localization and translational regulation of the grk message. We show that Cup and polyA-binding protein (PABP) interact physically with Sqd and with each other in ovaries. We show that cup mutants lay dorsalized eggs, enhance dorsalization of weak sqd alleles, and display defects in grk mRNA localization and Grk protein accumulation. In contrast, pAbp mutants lay ventralized eggs and enhance grk haploinsufficiency. PABP also interacts genetically and biochemically with Encore. These data predict a model in which Cup and Sqd mediate translational repression of unlocalized grk mRNA, and PABP and Enc facilitate translational activation of the message once it is fully localized to the dorsal-anterior region of the oocyte. These data also provide the first evidence of a link between the complex of commonly used trans-acting factors and Enc, a factor that is required for grk translation.

Figure 1. Cup and PABP55B interact with Sqd in ovarian extracts

(A) Immunoprecipitations were performed out of ovarian lysates using either α-Sqd or a negative control antibody, α-Dorsal. Specific Sqd interactors (marked by arrows) were excised from the gel and sequenced by mass spectrometry. Positively identified bands are labeled. (B-E) Immunoprecipitations were performed in the presence or absence of RNase (if indicated) using α-Sqd, α-Cup, α-PABP55B, or α-Dorsal. The lysate lane represents 5% of the sample probed for interactions. Western blots were probed with α-PABP55B (B and E) or α-Cup (C-E). These experiments demonstrate a strong interaction between Sqd, Cup, and PABP55B. (E) Additional immunoprecipitations were performed including α-Hrb27C/Hrp48 and probed with α-Cup and α-PABP55B. In addition to the regular exposure of the Cup western, an extended exposure is shown to clearly demonstrate the somewhat weaker interaction of Hrb27C/Hrp48 with Cup.

Figure 2. cup and pAbp55B females lay eggs with dorsoventral patterning defects

Most pAbp^EY11561 / pAbp^k10109 and pAbp^EY11561 eggs display wild-type morphology (A), but approximately 4% of eggs laid by pAbp^EY11561 / pAbp^k10109 females and 8% of eggs laid by pAbp^EY11561 homozygous females (B) exhibit partially or completely fused dorsal appendages. A small percentage of cup⁵ eggs (n=135) have wild-type morphology (A), but approximately 95% of eggs laid by cup⁵ / cup⁵ females have thick, fused dorsal appendages (C), a crown of dorsal appendage material (D), or open chorions (E).

Figure 3. Grk protein is not properly localized in cup mutants

OregonR, cup⁵, and cup^Δ212 egg chambers were stained for Grk (red), F-actin (green), and DNA (blue). Grk protein distribution was categorized as either dorsal-anterior only (A), dispersed throughout the oocyte with a dorsal-anterior bias (B), or evenly dispersed throughout the ooplasm (C). +/− indicates standard deviation of multiple microscopy sessions. Binomial probabilities for the frequency of each classification in cup mutants were calculated relative to OregonR. *, p < 0.05; ** p < 0.01.

Figure 4. grk mRNA is localized less efficiently in cup mutants

In situ hybridization using a grk RNA probe was performed on OregonR, cup⁵, and cup^Δ212 egg chambers. grk mRNA distribution was categorized as either dorsal-anterior only (A), an anterior ring with a dorsal-anterior bias (B), or an anterior ring (C). +/−indicates standard deviation of multiple microscopy sessions. Binomial probabilities for the frequency of each classification in cup mutants were calculated relative to wild type. †, p = 0.16; **, p < 0.01.

Figure 5. sqd interacts genetically with cup, and grk interacts genetically with pAbp

Heterozygosity for cup²⁰ enhances the moderately dorsalized phenotype of sqd^k12 / sqd¹ transheterozygotes and of sqd^k12 / sqd^k12 at 29°C (A). Eggs were characterized as either wild-type like, weakly dorsalized (single, broad fused appendage), moderately dorsalized (widely-spaced appendages), or strongly dorsalized (crown of appendage material). In addition, grk^HF48 / pAbp^k10109 transheterozygotes lay an increased percentage of and more severely ventralized eggs at 25°C than does either heterozygote alone (B). Eggs were characterized as either wild-type like, weakly ventralized (appendages fused at the base), moderately ventralized (single, slender fused appendage), or strongly ventralized (no appendage material). +/− indicates standard deviation of multiple egg collections scored by independent investigators. Binomial probabilities for the frequency of each eggshell classification were calculated for cup²⁰/+ ; sqd^k12 / sqd¹ relative to sqd^k12 / sqd¹, cup²⁰/+ ; sqd^k12 / sqd^k12 relative to sqd^k12 / sqd^k12, and grk^HF48 / pAbp^k10109 relative to grk^HF48 / +. All p-values were less than 0.01.

Figure 6. pAbp55B interacts with Enc biochemically and genetically

(A) Immunoprecipitations were performed out of ovarian lysates in the presence or absence of RNase using α-Enc or α-Dorsal. The lysate lane represents 5% of the sample probed for interactions. Western blots were probed with α-PABP55B. (B) Heterozygosity for pAbp^k10109 enhances the weakly ventralized phenotype of enc^Q4 / enc^Q4 and of enc^UU3/enc^UU3 homozygotes at 25°C, increases the percentage of collapsed eggs, and decreases the overall number of eggs laid. Eggs were characterized as either wild-type like, weakly ventralized (appendages fused at the base), moderately ventralized (single, slender fused appendage), strongly ventralized (no appendage material), or collapsed. Binomial probabilities for the frequency of each eggshell morphology classification were calculated for enc^Q4 / enc^Q4 ; pAbp^k10109 / + relative to enc^Q4 / enc^Q4 mutants and enc^UU3 / enc^UU3 ; pAbp^k10109/+ relative to enc^UU3 / enc^UU3 mutants. †, p > 0.05; *, p < 0.05; **, p < 0.01.

Figure 7. Model for the role of Sqd, Cup, PABP55B, and Enc in grk expression

Before grk mRNA is fully localized to the dorsal-anterior region of the oocyte, translational repression is mediated by Sqd, Hrb27C/Hrp48, Otu and Cup. Once the RNA is localized, PABP55B mediates translational activation of the localized message. Enc, bound to PABP55B, could function in association with a cytoplasmic anchor to mediate the transition from translational repression to activation of grk mRNA.