To better understand embryonic function of Casp, we decided to first characterize a casp allele, w¹¹¹⁸; pBac(PB)casp^c04227 (Bloomington stock number:11373; Figure 1—figure supplement 1A). This allele is induced by a piggyBac insertion in the 5’ regulatory region of CG8400/FBgn0034068/Casp locus which is situated at 2 R, region 52D14-15 (Thibault et al., 2004). As summarized in Figure 1, our data indicate that casp^c04227 is a strong hypomorphic allele and is referred to as a ‘loss of function’ allele (casp^lof) here onwards.

To better understand the function of Casp, we first decided to analyze how casp RNA and Casp protein are expressed during embryogenesis. Consistent with the maternal effect lethality, modENCODE (Muers, 2011; Chen et al., 2014) data suggests that casp mRNA is deposited maternally, being highly expressed in the 0–2 hr embryo, with a drop in expression 2–4 hr post fertilization (Figure 1—figure supplement 1B; Brown et al., 2014). Publicly available in-situ databases such as FlyAtlas, Fly-FISH, and BDGP (Weiszmann et al., 2009; Tomancak et al., 2002; Lécuyer et al., 2007) also confirmed maternal deposition of casp mRNA, with ubiquitous expression seen in stage 1–3 embryos. Fly-FISH data (Lécuyer et al., 2007; Wilk et al., 2016) further suggests that expression in the pole cells persists while maternally deposited somatic transcripts are selectively degraded from stage 4 onwards. We immunostained 0- to 3-hr-old embryos derived from w¹¹¹⁸ mothers with the anti-Casp antibody (Tendulkar et al., 2022) and found that Casp protein was predominantly cytoplasmic, and as in the case of RNA, exhibited a relatively ubiquitous distribution in the somatic compartment (Figure 1D). Again, consistent with the in situ hybridization pattern, Casp-specific antibody staining appeared to be enriched in the posteriorly localized PGCs compared to the surrounding soma (Figure 1D, arrow). To establish the specificity of the anti-Casp antibodies in the embryonic context, we stained the embryos laid by both the casp^lof/ casp^lof and casp^lof/Df mothers. As expected, Casp protein was nearly absent in embryos of both these genotypes suggesting that Casp protein is significantly reduced in the 0- to 3-hr-old casp^lof mutant embryos. Analysis of the embryonic lysates using western blotting, suggested that trace amounts of Casp protein (estimated 5–10% of w¹¹¹⁸), were present in embryos laid by casp^lof/casp^lof and casp^lof/Df mothers, supporting the conclusion that casp^lof is a strong hypomorphic allele and not a null (Figure 1E). Interestingly, mature egg chambers of wild-type animals had Casp expression in somatically derived follicle cells, but Casp protein could not be readily detected in the egg itself (Figure 1—figure supplement 1C). This observation suggested that, unlike the RNA, only trace amounts of Casp protein may be maternally deposited. Thus, the protein detected in the young embryos is likely generated post-fertilization, by the translation of maternally deposited casp mRNA.

Larval cuticle patterns are an excellent readout for major patterning defects in Drosophila (Nüsslein-Volhard and Wieschaus, 1980). To better understand the function of Casp protein during embryonic development, we analyzed the cuticles of casp mutant embryos. 0- to 3-hr-old embryos laid by casp^lof/Df mothers were collected and aged for 22 hr to prepare cuticles (see Materials and methods). The cuticular patterns were visualized under a dark field microscope. While ~60% of larvae displayed cuticular patterns comparable to wild type larvae, development in the remaining 40% embryos appeared to have stalled before they could deposit cuticle. These data were consistent with the extent of maternal effect lethality. Next, to define the stage of developmental arrest, we performed time-lapse live Bright-field imaging (Cavey and Lecuit, 2008). These data (Movies, Figure 1—videos 1–3) confirmed that 40% of embryos did not proceed to gastrulation and showed developmental arrest before germ band extension (around stage 6; Figure 1—figure supplement 2 B4, C4). Such embryos displayed irregular, uncoordinated morphogenetic movements with blebbing of the plasma membrane possibly inhibiting both cephalic furrow formation, and germband elongation (Figure 1—figure supplement 2 B2, B3, C2, C3).

To trace the defects during mid-embryogenesis including failure of gastrulation, we sought to visualize casp^lof embryos at blastoderm stage. We labelled the 0- to 3-hr-old casp^lof and control embryos with the nuclear dye Hoechst and the cytoskeletal F-actin marker phalloidin (Figure 2A and B). casp^lof embryos displayed significant structural abnormalities that are reflected in an irregular actin network that lacks stereotypical organization, which is partially interrupted in several places (Figure 2 B1 compared to Figure 2 A1). Similarly, unlike the age-matched control embryos, regular nuclear spacing is disrupted, and nuclei are unevenly distributed across the embryo with occasional instances of nuclear fusion and possibly mitotic asynchrony (Figure 2B vs Figure 2A).

The nuclear and cytoskeletal defects observed in casp embryos prompted us to analyze the centrosomes in embryos deficient in casp function. Centrosomes function as the microtubule organizing centers, and actomyosin-based cytoskeletal defects have been correlated with aberrant centrosome activity (Wu and Akhmanova, 2017; Blake-Hedges and Megraw, 2019 and references therein). In wild-type nuclei, centrosome duplication occurs simultaneously with the initiation of the nuclear division cycle. After completion of duplication, centrosomes separate and migrate along the nuclear membrane to reach the opposite sides of the nucleus (reviewed by Lattao et al., 2017; Wu and Akhmanova, 2017; Blake-Hedges and Megraw, 2019). In the embryos maternally compromised for casp, several characteristic aspects of centrosome behavior during mitotic divisions are altered (compare Figure 2 B3 to Figure 2 A3). At times, centrosomes appeared to divide even without a nucleus (or DNA), and many ‘orphan’ centrosomes devoid of nuclear DNA were observed (Figure 2 B3, inset). By contrast, in some instances, centrosomes were duplicated but remained in proximity indicating failed or incomplete migration (see arrows in Figure 2 B3, inset). Quantitation of the centrosomal abnormalities in casp^lof embryos (Figure 2C and D) further revealed that the behavior of the centrosomes, especially their ability to separate and migrate to the opposite poles correctly deteriorated progressively (compare Figure 2D to Figure 2C), as nuclear division cycles advanced.

To confirm the specificity of the phenotypic consequences induced by the maternal loss of casp, we overexpressed UAS-casp^wt transgene in casp^lof/Df embryos using nos Gal4, a maternal Gal4 driver. Embryos derived from such mothers were stained using nuclear dye Hoechst, phalloidin and anti-Gamma tubulin antibodies that mark the centrosomes. As shown in Figure 2E, maternal overexpression of casp substantially rescues centrosomal and cytoskeletal abnormalities seen in casp^lof (compare panels E1-E3 to A1-A3). This rescue is quantitated in Figure 2G and H (compare with Figure 2C and D). The casp^lof/Df; nos Gal4/UASp-casp^wt showed a rescue of lethality of 20–25% as compared to 40–45% for casp^lof/Df.

Taken together, these data showed that Casp plays an essential role in early embryonic development in Drosophila, and loss of casp results in conspicuous nuclear and cytoskeletal defects that correlate with incomplete gastrulation movements and developmental arrest, for ~45% of embryos. Moreover, such defects do not appear to be region-specific or localized within an embryo arguing in support of ubiquitous function across the embryo for Casp protein.

Primordial germ cells (PGCs) are precursors of the gametes—sperm and eggs. Their specification is essential for the development of the germline, which in turn passes genetic information to the next generation. During early embryonic development, PGCs are specified and segregated from somatic cells. In Drosophila, the PGCs are formed in the posterior end of the embryo. As the early embryonic syncytial nuclear division cycles progress, a few nuclei and centrosomes associated with them invade posteriorly localized and anchored pole plasm ahead of the rest of the somatic nuclei (Raff and Glover, 1989). The precocious entry of the centrosomes results in the release and microtubule-dependent transport of pole plasm, which is sequestered in newly cellularized PGCs (Raff and Glover, 1989; Lerit et al., 2017). Thus, the PGCs or the germline stem cell precursors are set aside during early embryogenesis. Moreover, centrosome behavior and dynamics are crucial for proper cellularization and mitotic cell divisions of early germ cells (Lerit and Gavis, 2011; Lerit et al., 2017). Centrosomes, the microtubule (MT)-organizing centers, ensure the faithful segregation of germ plasm, a reservoir of germ cell determinants, into PGCs. Taken together with the observation that Casp protein is readily detectable in early PGCs, we wondered if it could influence the total pole cell count. To examine this possibility, stage 5 embryos derived from cs and casp^lof/ Df mothers were immunostained for the pole cell marker Vasa and visualized under a confocal fluorescence microscope. Interestingly, while cs and nos-Gal4 embryos had an average of ~30 pole cells (Figure 3 A1-B3, for quantitation see panel F) casp^lof/ Df showed a drastic reduction in total PGC count to ~10 (Figure 3 C1-3, F). By contrast, stage 5 embryos derived from nosGal4 >UAS casp mothers, where casp was overexpressed (Figure 3G), displayed considerably elevated total number of PGCs (Figure 3 E1-E3, F). To confirm that the loss of casp is specifically responsible for the reduction in the total number of PGCs, we simultaneously overexpressed casp in the casp^lof/Df mothers in a germline-specific manner, which resulted in a significant rescue (~26 pole cells; Figure 3 D1-D3, F). Taken together, these data suggested that the total pole cell number in late syncytial /cellular blastoderm embryos specifically depends on casp function. Furthermore, increasing casp levels maternally, is sufficient to elevate the total pole cell number substantially.

Data presented in the previous section demonstrated that both the ‘loss’ and ‘gain’ of casp activity exert reciprocal influence on total PGC numbers in early embryos. As very few germ plasm components have been shown to display this trait (Jongens et al., 1994), we decided to explore the phenomenon further. We have previously reported that Casp protein physically associates with both transitional endoplasmic reticulum ATPase (TER94; also called valosin-containing protein, VCP or p97) and vesicle-associated membrane protein-associated protein B (VAPB/VAP33A)(Tendulkar et al., 2022). Furthermore, based on interaction studies and biochemical analysis, we proposed that Casp may act as an adapter that either directly or indirectly mediates the physical association between VAPB and TER94 (Tendulkar et al., 2022). To arrive at this conclusion, we had primarily relied on S2 cell lysates as well as adult and fly head total protein extracts. To specifically identify the major protein partners of Casp in the early embryos, we performed similar immunoprecipitations using anti-Casp antiserum in early (0–3 hr) embryonic lysates and interactors were identified via mass spectrometry (Figure 3—figure supplement 1A). Peptides from 122 proteins were recovered in the Casp IP but not the IgG IP and thus, were considered significant interactors. The top 23 interactors are presented in Figure 3—figure supplement 1A. Interestingly, as in the case of brain and S2 cell lysates, TER94 and VAPB were the top hits, with germ cell determinants also identified as interactors (Figure 3—figure supplement 1C; Discussed in a subsequent section). These observations suggested that a functional protein complex between Casp, TER94, and VAPB likely exists in many different tissue/cellular contexts, including early embryos.

Mammalian VCP/p97 is an essential chaperone for proteostasis that modulates several ubiquitin-associated processes (Peters et al., 1990; Dai et al., 1998; Meyer et al., 2000; Meyer, 2005; Ye, 2006; Jentsch and Rumpf, 2007; Meyer et al., 2012). Its Drosophila counterpart TER94 has been studied mostly in the context of neurodegeneration (Griciuc et al., 2010; Chang et al., 2011; Azuma et al., 2014; Kushimura et al., 2018; Tendulkar et al., 2022; Thulasidharan et al., 2024). Loss of TER94 was shown to ameliorate polyQ-induced eye degeneration. Moreover, the overexpression of TER94 promoted the apoptosis of neuronal cells (Higashiyama et al., 2002). TER94 has been uncovered in a genetic screen for maternal proteins that are phosphor-regulated (Zhang et al., 2018) and have roles in oogenesis and early development.

TER94 is one of the significant interactors of Casp protein in the embryonic context as suggested by the Casp interactome (Figure 3—figure supplement 1A) and has been implicated in ER-associated degradation (Chang et al., 2011; Tendulkar et al., 2022). Thus, we wondered if TER94 also has a maternal function and whether embryos maternally compromised for TER94 display similar phenotypes as casp. To achieve this, we employed VALIUM 20/22 maternal RNAi lines (UAS-TER94i). First, we used a mat-α4tubulin:VP16-Gal4 (Mat-αtubGal4) driver to deplete TER94 activity in the late stages of oogenesis (Figure 4).

We sought to observe the consequence of TER94 knockdown both in the adult females and early embryos derived from such females. In the females compromised for TER94, egg-laying behavior remained largely unaffected. Interestingly, however, the viability of such embryos was severely impaired, with >95% of embryos failing to hatch (Figure 4A), in agreement with earlier studies (Zhang et al., 2018).

A western blot of lysates derived from Mat-GAL4 >UAS-TER94i mothers probed with anti-TER94 antiserum indicated a robust knockdown of TER94 as compared to control (Figure 4B). Phenotypic analysis of embryos derived from Mat-GAL4 >TER94 i females revealed that >70% of Mat-GAL4 /TER94 i embryos failed to progress to the syncytial blastoderm stage, with even fewer reaching Stage 5. The late-stage syncytial/ cellular blastoderm embryos were further assessed for cell cycle defects. As in the case of casp^lof/Df, Mat-GAL4 >TER94 i embryos also displayed defects ranging from irregular nuclear distribution to perturbed F-actin localization, as assessed by HOECHST and phalloidin staining, respectively (compare Figure 4 D1–D3 with Figure 4 C1–C3).

We also examined if maternal loss of TER94 activity results in centrosome aberrations like those observed in casp^lof embryos. Indeed, embryos deficient in TER94 function showed both the characteristic phenotypes observed in the casp embryos including (a) inefficient separation of duplicated centrosomes and (b) multiple instances of ‘orphan’ centrosomes devoid of nuclear DNA (Figure 4E, F). An earlier study (Zhang et al., 2018) had found that TER94 RNAi leads to multi-polar spindles with supernumerary centrosomes.

Taken together, these data demonstrated that embryos maternally compromised for either Casp or TER94 functions share several phenotypes, suggesting that both the proteins likely perform essential and likely related functions during syncytial nuclear cycles in Drosophila embryos. Furthermore, the phenotypic consequences observed upon maternally compromising casp and TER94 overlap but are not identical, with TER94 embryos showing higher penetrance in terms of phenotypes and embryonic lethality. Future experiments will be necessary to resolve the functional distinction between the two.

Maternal loss of TER94 function mimicked both the cytoskeletal and corresponding centrosome aberrations observed upon similar loss of casp activity. Both centrosomes and germ plasm, are essential for the proper specification and formation of PGCs in Drosophila embryos (Lerit and Gavis, 2011; Lerit et al., 2017). This prompted us to examine the Casp interactome for the possible enrichment of germ plasm components and proteins that may regulate centrosome dynamics and/or behavior. Figure 3—figure supplement 1B lists Casp interactors that are important constituents of the germplasm, including eIF4A, me31B, and TER94, known constituents of the polar granules (Thomson et al., 2008).

Among the different components of the germ plasm, Oskar serves as the principal determinant of the PGC fate (Kim-Ha et al., 1991; Snee and Macdonald, 2004; Vanzo et al., 2007; Lehmann, 2016). Supporting the conclusion, the loss and gain of function of oskar leads to reciprocal phenotypes. Compromising oskar activity maternally, leads to reduction in total number of PGCs whereas anterior ectopic expression of oskar using the bicoid mRNA localization signal induces pole cell formation at the anterior (Ephrussi and Lehmann, 1992). The germ plasm is supplemented with mitochondria and polar granules, which consist of ribonucleoprotein complexes. Downstream of oskar, vasa and tudor are two genes that are essential for the assembly of pole plasm (Breitwieser et al., 1996; Arkov et al., 2006; Jones and Macdonald, 2007). Biochemical proteomic analysis of Vasa (VAS) and Tudor (TUD) containing polar granule complexes identified eIF4A, me31B, and TER94. This indicated that the germ plasm consists of the components of translational machinery and the endoplasmic reticulum assembly (Thomson et al., 2008).

TER94 is a component of pole plasm and is also physically associated with Casp. Furthermore, proteomic analysis indicated that Casp could be part of a complex comprising several germ plasm proteins including Tudor and Me31B. To confirm that, as in the case of Casp, TER94 is also detected in PGCs. We used antibodies generated against Casp (Figure 3—figure supplement 1D) and TER94 (Figure 3—figure supplement 1E) to label wild-type embryos. Embryos were also co-immunostained for germ cell specific marker, Vasa. As can be seen, both Casp and TER94 proteins are found in PGCs, colocalized with Vasa.

As we were specifically interested in investigating possible similarities and distinctions between the respective functions of TER94 and Casp during early embryonic germ cell development, we decided to examine their possible functions during PGC formation (also referred to as pole cell budding). This seemed especially pertinent as centrosome behavior and dynamics are crucial for proper formation and cellularization of PGCs. Evidently, a germ plasm component, Germ-cell-less (Cinalli and Lehmann, 2013) was shown to be necessary for proper separation of daughter centrosomes in the dividing pole buds (Lerit et al., 2017). Similarly, loss of centrosome components such as Centrosomin results in partial loss of early embryonic PGCs. As maternal loss of both casp and TER94 results in aberrant centrosomes in the surrounding somatic nuclei in early blastoderm embryos, we sought to investigate if posteriorly positioned centrosomes in the vicinity of pole plasm also display similar problems. Indeed, maternally compromising casp led to significant loss of pole buds and corresponding defective centrosome behavior (Figure 5).

Especially these pole buds display defective centrosome separation and/or orphan centrosomes (Figure 5 B1–B2, D and E), as observed in the somatic nuclei. Intriguingly, while casp^lof embryos show discernible loss of pole buds, TER94i embryos displayed an increased number of round-shaped, ‘bud-like’ cells at the posterior of the stage 3 embryos (Figure 5 C1–C2). Typically, such ectopically localized pole buds also showed inadequate separation of centrosomes and could be identified as pole buds due to the enrichment of Vasa. However, Vasa distribution and accumulation in TER94i buds was variable and non-uniform. Since TER94 is required for Osk mRNA localization in the oocyte, the defects in pole cell budding including their ectopic positioning could be a consequence of inappropriate segregation of oskar. Moreover, the early nuclear division cycles and nuclear migration defects seen in are defective in TER94i embryos could also contribute to PGC formation/ cellularization. Notably, although the pole bud count is elevated in TER94i embryos, presumably, these pole buds don’t survive through subsequent nuclear cycles, and consequently, late blastoderm TER94i embryos almost completely lack PGCs.

Taken together, our data demonstrate that change in Casp levels leads to corresponding alteration in the total PGC count. Moreover, PGC formation is affected by maternal loss of Casp, although the precise nature of its involvement in this process remains to be determined. As Oskar is the master determinant of germ cell fate, we sought to determine if Oskar levels (Kim-Ha et al., 1991; Ephrussi and Lehmann, 1992) are correspondingly altered upon change in Casp. To assess this possibility, we stained wild-type control embryos and embryos maternally compromised for casp using anti-Oskar antibodies (compare Figure 6B to Figure 6A). Simultaneously, we stained embryos derived from nosGal4 >UASp casp mothers (Figure 6 C1–C3) that display significantly elevated number of PGCs, presumably due to excess levels of Casp protein in the germline.

Satisfyingly, the change in Oskar protein levels is consistent with the alteration in the Casp levels in the maternal germline (Figure 6D). Embryos derived from casp mutant mothers show considerably diminished levels of Oskar as compared to the control whereas overexpression of Casp in the maternal germline results in elevation in the Oskar protein amount (Figure 6D). Furthermore, in all instances, Oskar protein appears to be anchored to the posterior pole as the wild-type control embryos and the only discernible change is observed in its levels.

Changes in Casp protein levels influence the accumulation of Oskar protein. Oskar is necessary and sufficient to assemble pole plasm at the posterior pole which controls the total number of pole buds in an embryo. To directly evaluate the influence of Casp on pole bud formation, total number of pole buds and pole cells were quantitated upon maternal loss and gain of Casp against the control embryos at the same stage (Figure 6E). Consistent with the data presented in the previous sections, total number of pole buds (nc 9 and 10) were reduced in casp^lof embryos whereas the number was elevated in the presence of excess Casp (Figure 6E). Similar quantitation was performed in older embryos in two stages (nc11-12 and NC12-13), and a progressive increase was observed in the total number of PGCs upon the gain of Casp function (Figure 6E). By contrast, the total PGC count for casp^lof/Df embryos did not increase appreciably at nc11-12 and nc12-13.

Primordial germ cell number is determined by limited mitotic cell divisions that each pole bud undergoes in a stochastic and asynchronous manner. As these are mitotic divisions, the dividing PGCs can be identified using antibodies against the miotic marker phosphoHistone3 (pH3). As the total number of pole buds and PGCs change per Casp levels, we wanted to examine if this is also reflected in the total number of pH3-positive germ cells simultaneously labeled with the anti-Vasa antibodies. We decided to focus on syncytial blastoderm embryos between NC12-14. As anticipated, maternal overexpression of Casp led to a corresponding increase. In control (nos gal4) embryos, roughly 20% PGCs display either 1 or >1 pH3 positive PGCs whereas, 70% of nos-gal4 >Casp^WT embryos show 1 or >1 pH3 positive PGCs (Figure 6F). Conversely, less than 10% of casp^lof embryos showed either 1 or >1 pH3 positive PGCs and pole cells with 3 or >3 pH3 positive cells were completely absent in this background as opposed to 25% present in the control embryos (Figure 6F). Importantly, maternal expression of the rescue construct in the casp^lof embryos ameliorated the loss of pH3-positive PGCs seen in just the mutant (casp^lof) PGCs.

Our analysis thus far has revealed two related yet distinct phenotypes associated with maternal loss of casp. First, it can influence proper PGC formation and specification during early embryonic development via its effect on the accumulation of germ cell determinant, Oskar. Subsequently, it helps orchestrate cellular movements leading up to gastrulation during mid-embryogenesis. While the first activity is likely germ cell autonomous, the second relates to its role in the somatic cells/nuclei. Moreover, both the somatic and germline compartments of casp^lof embryos share centrosomal as well as cytoskeletal aberrations. We thus wondered whether the two activities are mechanistically connected, and if the possible connection relates to the (MZT). One aspect of the MZT that has been highlighted in recent years is the active degradation of maternal proteins (Cao et al., 2020). These proteins comprise 2% of the maternal proteome and are degraded abruptly, at the end of the MZT. In Drosophila, ubiquitin-proteasome-based degradation of three repressor proteins, namely Smaug (Smaug), Trailer hitch (Tral), Maternal expression at 31B (Me31B), marks the MZT (Cao et al., 2020). We thus decided to probe, using antibodies generated against these proteins (Figure 7) if the degradation of any of these proteins is affected due to maternal loss of Casp/casp during the initial hours of embryonic development (0–1, 1–2, and 2–3 hr). TER94 protein is unchanged (Figure 7 D1, D2) as is the α-tubulin loading control (Figure 7 E1, E2). Furthermore, pattern of degradation of Me31B and TRAL proteins in the casp^lof embryos is relatively unaffected, when compared to the w¹¹¹⁸ control samples.

By contrast, Smaug (Figure 7 C1, C2) is an interesting exception. In control, 0- to 1-hr-old embryos, low levels of Smaug protein were observed, which increased in the 1- to 2-hr-old embryo extract but decreased the 2–3 hr time-window (Figure 7C), data that agrees with previously published work (Cao et al., 2020). However, in the casp^lof/Df embryos levels of Smaug protein are modestly elevated from the start (0–1 and 1- to 2-hr-old extract samples respectively; Figure 7 C2). Critically, unlike control embryos, Smaug protein persists in 2- to 3-h r-old casp^lof/Df embryonic extracts (Figure 7 C2, *). Altogether, these data argue that Smaug degradation is specifically adversely influenced in the casp^lof/Df embryos at the end of the MZT.

Since Smaug levels appear to be elevated in embryonic lysates in the MZT, we decided to test if the Smaug levels are elevated specifically in casp^lof/Df pole cells. To this end, pre-syncytial as well as early syncytial blastoderm embryos of both the genotypes (control and casp^lof/Df) were stained with anti-Smaug and anti-Vasa antibodies. Aligned with the Western blot data presented earlier, levels of Smaug protein are significantly elevated in pre-syncytial embryos (Figure 8, compare panel B3 with A3). Also, in casp^lof/Df embryos, Smaug protein seems to accumulate in discernible large puncta which are barely visible at this stage in control embryos (Figure 8, B3 vs A3). At the syncytial blastoderm stage (Stage 4), however, the Smaug-positive puncta appear larger and more numerous in the few surviving PGCs from the casp^lof/Df embryos (Figure 8 D3) when compared to controls (Figure 8 C3).

Taken together, our data suggest that Smaug protein levels are appreciably elevated in pole cells from casp^lof/Df embryos. The specific increase in Smaug levels may, in part, be due to inappropriate accumulation and/or stabilization of Smaug. Smaug protein is necessary for translational control of the posterior determinant nanos. Early reports indicated that unlocalized nos RNA is translationally repressed by Smaug which binds to Smaug response elements (SREs) within 3’UTR of nos RNA. In addition to nos, it also regulates hsp83 translation. Smaug participates in multiple, overlapping mechanisms including interaction with the components of the translation machinery as well as deadenylation to regulate translation/localization of the target RNAs (Dahanukar et al., 1999; Zaessinger et al., 2006).

Intriguingly, recent data from Lipshitz lab has, in fact, shown that Smaug protein accumulates in germ granules (Siddiqui et al., 2023). Moreover, in addition to its canonical role in regulating nos translation, it can also repress oskar at the translational level (see Discussion). As the embryonic PGC count depends upon Oskar levels, Smaug mutant embryos show an increase in the total number of PGCs. These data fit nicely with our observations and are entirely consistent with a model incorporating the downregulation of Smaug mediated by Casp.

To decipher the specific functions of the individual protein domains of Casp during pole cell formation and division we followed an experimental strategy involving rescue of the loss-of-function phenotype. We compared the extent of rescue observed in the presence of individual deletion constructs to the full-length wild-type control. To this end, fly lines expressing domain deletions of Casp under the pUASp promoter were generated (Figure 9A). The pUASp-casp^ΔUBA, pUASp-casp^ΔUBX and pUASp-casp^ΔUAS construct represent deletions of the UBA, UBX, and UAS domains, respectively. Additionally, we also generated fly lines that express the pUASp-casp^ΔUASΔUBX which deletes both the N and C-terminal domains simultaneously. A full-length coding sequence of Casp was also cloned, and this pUASp-casp^WT line was used as a positive control carrying the ‘rescue’ construct. A western blot confirmed the expression of these constructs in adult animals with different domain deletions (Δ), UAS-casp^Δ driven by nos-GAL4 driver (Figure 9B). To assess the domain-specific functions of Casp in isolation without the confounding effects of endogenous Casp, the fly lines were balanced with a casp^lof allele on the second chromosome. These balanced lines were crossed to a fly line with the maternal driver nos-GAL4 on the third chromosome and an allele with a deficiency for casp (Df). Embryos laid by mothers of the genotype casp^lof/Df; nos-GAL4/pUASp-casp^Δ were used to determine maternal effects of the domain deletions (Figure 9C).

Embryos were collected from females of different genotypes and stained using anti-Vasa antibodies. As can be seen by the comparison shown in the bar graph (Figure 9C), in terms of the rescue of PGC numbers, deletion of the UBA domain does not affect the rescuing activity of the Casp protein, the UBX domain deletion can rescue, but is weaker than the UBA domain deletion, whereas deletion of the UAS domain is the least effective with regards to the rescue.

To extend these observations, we overexpressed individual Casp deletion mutants in an otherwise wild-type background and counted the total number of PGCs. Neither the maternal overexpression of pUASp-casp^wt nor any of the pUASp-casp^Δ increased lethality in terms of decrease in hatching of embryos, in all cases the hatching was in the range 90–95%. Again, as seen before, overexpression of casp^wt under these conditions led to a significant increase in the total number of PGCs in blastoderm embryos, with average PGC’s at 40. The ability to enhance the PGC division is retained in the absence of the UBA domain as, in this particular background, the mean count of the total number of germ cells was similar to embryos that expressed full-length protein. By contrast, deletion of the UAS domain is unable to support the (increased) germ cell division, and the total PGC count dropped to the same level as nos Gal4 embryos. Interestingly, deletion of UBX domain, yet again, displayed intermediate activity, with the ΔUBX-ΔUBA domain deletion also being unable to drive the increase in PGCs.

Proteomic analysis of the embryonic extracts suggested that in addition to TER94 protein, Casp is also associated with VAP (Figure 9—figure supplement 1A; Tendulkar et al., 2022). We thus sought to test if physical interaction between Casp and VAP is relevant for pole cell formation. To assess this possibility, we used a deletion variant of Casp, Casp^∆FFAT, which cannot interact with VAPB (Tendulkar et al., 2022). Expression of casp^∆FFAT under the control of maternal driver, in a casp^lof/Df mother showed hatching (~60%) frequencies at par with wild type (Figure 9—figure supplement 1B), with expression levels equal to other casp constructs (Figure 9—figure supplement 1C). Importantly, nos-Gal4-dependent maternal overexpression of UAS-casp^∆FFAT led to an increase in the total number of PGCs in syncytial blastoderm embryos, comparable to full-length UAS-casp^wt (average PGC ~23). Lastly, expression of casp^∆FFAT could rescue the PGC loss upon maternal loss of casp (compare Figure 9—figure supplement 1 F1-3 to Figure 9C). These data suggested that interaction between Casp and VAP is either not essential or partially redundant for Casp function in the context of determining PGC. Altogether, the functional analysis of different protein domains within Casp underscores the importance of UAS and UBX domains, especially in embryonic contexts, including PGC development.

Flies were raised on standard cornmeal agar at 25 °C unless stated otherwise. Casp^lof (11373), Casp Df (23691), nos Gal4 (4937), Mat α4-tubulinGal4:VP16 (7063) and TER94 RNAi (32869) lines were procured from the Bloomington Drosophila Stock Centre, with numbers in brackets indicated Stock numbers.

UASp-casp^Δ constructs (pUASp casp^ΔUBA, pUASp casp^ΔUAS, pUASp casp^ΔUBX, pUASp casp^ΔUASΔUBX) were generated, starting from a pUASp-attB-Casp construct (Tendulkar et al., 2022), with the design including an N-terminal HA tag. The constructs were injected into a w¹¹¹⁸; Attp2 embryo in the NCBS-CAMP transgenic injection facility, and stable lines were generated by balancing against a w¹¹¹⁸; TM3Sb/TM6Tb animal.

For cloning and amplification, a functional N-terminal HA tag was introduced with the 5’-forward primers and a 3’-homology arm was introduced using the 3’-reverse primers. The exception was the pUASp Casp^ΔUAS construct where two fragments, one upstream and one downstream of the UAS domain were amplified separately with 22 nucleotide homologous overhang between them to mediate homologous recombination. The upstream fragment contained the HA sequence while the downstream fragment contained the 3’-homology arm. The sequences of the primers used are as follows. TGTTCCAGATTACGCTGGCGGC was used as the 5’-forward primer for the pUASp casp^wt, pUASp casp^ΔUBX, pUASp casp^ΔUASΔUBX and the upstream fragment of the pUASp casp^ΔUAS constructs. TGTTCCAGATTACGCTGGCGGCCCCCATCCTATCCTGGTGCC was used as the 5’-forward primer for the pUASp casp^ΔUBA construct. CGGATGATGAGATAAGTGGCTCCACGGAAACATGCGAAATGTTTGAGGAGCAG was used as forward primer for amplifying downstream fragment of the pUASp casp^ΔUAS construct. ACCATGGGTTTAGGTATAATGTTATCAAGCTCC was the reverse primer for pUASp casp^wt, pUASp casp^ΔUBA, and downstream fragment of pUASp casp^ΔUAS. TTAGGTATAATGTTATCAAGCTCCTCATTCGGACGGCTCCTGAGGTAG was the reverse primer for pUASp casp^ΔUBX. TTAGGTATAATGTTATCAAGCTCCTCACGTGGAGCCACTTATCTCATCATCCG was used as the reverse primer for pUASp casp^ΔUASΔUBX. CGTGGAGCCACTTATCTCATCATCCG was the reverse primer for upstream fragment of pUASp casp^ΔUAS. PCR products were further amplified with a common 5’-primer ATAGGCCACTAGTGGATCTGATGTACCCATACGATGTTCCAGATTACGCTGGCGGC and their respective reverse primers to introduce the 5’-homology arm. Inserts were recombined into pUASp AttB vector linearised at the BamH1 site, using a variation of the SLiCE cloning method (Zhang et al., 2012; Zhang et al., 2014). In short, E. coli DH10B expressing the optimised λ-prophage red recombinase system (PPY cells) was cultured in the presence of arabinose to induce recombinase expression and subsequently used for preparing competent cells. DNA fragments containing homologous overhangs were co-transformed into the competent bacteria and resultant colonies were screened through PCR to identify proper recombinants and subsequently confirmed through sequencing.The pUASp casp^ΔFFAT was PCR amplified from pRM-HA:casp ^ΔFFAT (Tendulkar et al., 2022), cloned into pUASp-AttB, sequenced for validation and injected into a w¹¹¹⁸; Attp2 animal.

0–3 hr embryos were collected, transferred to a fresh sugar-agar plate, and unhatched larvae were scored after 48 hr to determine viability.

0–3 hr embryos were lysed in Co-IP Lysis Buffer (20 mM Tris pH 8.0, 137 mM NaCl, 1% IGEPAL, 2 mM EDTA, 1 X PIC) using a Dounce homogenizer, and centrifuged at 21,000 × g for 30 min. 3 mg of total lysate was incubated with 5 μg of primary antibody (Rb anti-Casp) and 5 μg of Normal Rabbit IgG overnight at 4 °C. Antigen-antibody complexes were captured using 50 μL of Bio-Rad SureBeads Protein A (1614013) at 4 °C for 4 hr. Beads were washed six times with Co-IP Lysis Buffer and protein complexes eluted by boiling in 1 X Laemmli Sample Buffer. Eluted proteins were resolved on a 10% polyacrylamide gel followed by western blotting or in-gel trypsin digestion, described in the following sections.

Embryos collected at varied time points (0–3, 0–1, 1–2, 2–3 hr) were lysed in RIPA buffer (50 mM Tris-Cl, 150 mM NaCl, 0.1% SDS, 0.01% Sodium azide, 0.5% sodium deoxycholate, 1 mM EDTA, 1% Triton X-100, 1 X PIC) with a pellet pestle (Kontes). Lysates were cleared by centrifugation at 21,000 × g at 4 °C for 30 minutes. Protein concentration was estimated using a BCA assay (Pierce) and 30–40 μg of total protein was loaded onto the gel after boiling in 1 X Laemmli Sample Buffer. Proteins separated by 10% SDS-PAGE were transferred onto a PVDF membrane (Immobilon-E, Merck) and blocked in 5% milk in Tris-Buffer Saline (TBS) with 0.1% Tween 20 (TBS-T) for an hour. Blots were then incubated overnight with primary antibody diluted in 5% milk in TBS-T, at 4 °C. Following three washes with TBS-T, blots were incubated with secondary antibodies diluted in 5% milk in TBS-T, for 1 hour at room temperature. Blots were washed thrice with TBS-T and visualized on a LAS4000 Fuji imaging system after incubating with Immobilon Western Chemiluminescent HRP substrate (Merck). The following antibodies were used: Rabbit anti-VAP, 1:10000 (Tendulkar et al., 2022), Mouse anti-α-Tubulin, 1:10000 (T6074, Sigma), Mouse anti-Ubiquitin, 1:1000 (P4D1, Santa Cruz Biotechnology), Mouse anti-HA, 1:5000 (H3663 HA-7, Sigma-Aldrich), Rabbit anti-HA, 1:2000 (04–902 DW-2, Sigma-Aldrich), Rabbit anti-Casp, 1:10000 (Tendulkar et al., 2022), 1:2000 (04–902 DW-2, Sigma-Aldrich). Goat anti-rabbit HRP, Goat anti-rat HRP and Goat anti-mouse HRP secondary antibodies, each at 1:10000 (Jackson ImmunoResearch). Rabbit anti-me31B (1:1000), Rabbit anti-Smaug (1:500), Rat anti-Tral (1:1000) were a kind gift from Elmar Wahle (Cao et al., 2020).

Before in-gel trypsin digestion of the Co-IP eluate, the antibody was crosslinked to the SureBeads using DMP (Sigma) according to the NEB crosslinking protocol to avoid elution of the antibody. After crosslinking 10 μg Casp antibody, Co-IP was performed as described above. In-gel trypsin digestion was carried out as previously described (Shevchenko et al., 2006). Briefly, Coomassie-stained bands on the gel were excised and cut into 1 mm cubes. Gel pieces were transferred to a clean microcentrifuge tube and destained with buffer containing 50% acetonitrile in 50 mM Ammonium bicarbonate. Reduction and alkylation were carried out on the destained gel pieces by incubating with 10 mM dithiothreitol (DTT) followed by incubating with 20 mM iodoacetamide. Gel pieces were saturated with sequencing grade Trypsin (Promega) at a concentration of 10 ng/μl and incubated overnight at 37 °C. Peptides were extracted by sequential addition of 100 μl of 0.4% Trifluoroacetic acid (TFA) in 10% ACN, 100 μl of 0.4% TFA in 60% ACN and 100 μl of ACN. The pooled extract was dried in a vacuum centrifuge and reconstituted with 50 μl of 0.1% TFA. The peptides in TFA were purified using the StageTip protocol (Rappsilber et al., 2007). LC–MS/MS analysis was performed on the Sciex TripleTOF6600 mass spectrometer interfaced with an Eksigent nano-LC 425. Tryptic peptides (1 μg) were loaded onto an Eksigent C18 trap (5 μg capacity) and subsequently eluted with a linear acetonitrile gradient on an Eksigent C18 analytical column (15 cm ×75 μm internal diameter). A typical LC run lasted 2 hr post loading onto the trap at a constant flow rate of 300 nl/min with solvent A consisting of water + 0.1% formic acid and solvent B consisting of acetonitrile. The gradient schedule for the LC run was 5% (vol/vol) B for 10 min, a linear gradient of B from 0% to 80% (vol/vol) over 80 min, 80% (vol/vol) B for 15 min and equilibration with 5% (vol/vol) B for 15 min. Data was acquired in an information-dependent acquisition (IDA) mode over a mass range of 300–2000 m/z. Each full MS survey scan was followed by MS/MS of the 15 most intense peptides. Dynamic exclusion was enabled for all experiments (repeat count 1; exclusion duration 6 s). Peptides were identified and quantified using the SCIEX ProteinPilot software at a false discovery rate (FDR) of 1%. A RefSeq Drosophila protein database (release 6) was used for peptide identification. Proteins that were identified in two or more replicates were tabulated.

0–3 hr embryos were collected in a sieve and dechorionated in 4% sodium hypochlorite for 90 s. After thorough washes with distilled water, embryos were fixed in a 1:1 heptane:4% PFA solution for 20 min. The PFA layer was removed, and embryos in heptane were re-constituted with an equal volume of methanol. Embryos were devitellinized by vigorous shaking in the 1:1 heptane:methanol mixture. The heptane layer and the interphase containing non-devitellinized embryos were carefully removed. Devitellinized embryos in the bottom methanol phase were washed twice with methanol and stored at –20 °C till they were ready to be imunostained. For phalloidin and Smaug staining, embryos fixed in heptane: 4% PFA were hand de-vitellinized, after which the standard immunostaining procedure was followed. For immunostaining, embryos were rehydrated by washing thrice with 0.3% PBS-TritonX 100 (PBS-T) for 15 min each. Embryos were blocked in 2% BSA in 0.3% PBS-T for 1 hr at room temperature (RT). Embryos were incubated at 4 °C overnight with primary antibodies diluted in 2% BSA in 0.3% PBS-T at the appropriate dilutions. Following three 15-min washes with 0.3% PBS-T, embryos were incubated in the appropriate secondary antibodies for 1 hr at RT. The following antibodies were used: Rabbit anti-Casp, 1:1000 (Tendulkar et al., 2022); Rat anti-α-vasa; 1:50 (DSHB); Rabbit anti-VCP, 1:200 (#2648, Cell Signaling Technology); Rabbit anti-Oskar, 1:1000 (generated by Mandy Jeske, Anne Ephrussi lab); Rabbit anti-Smaug, 1:50; Mouse anti-γ-tubulin, 1:1000 (T6557, Sigma), Rabbit P-Histone H3 (S10) 1:300 (#9701, Cell Signalling); Alexa Fluor 568 Phalloidin, 1:1000 (Invitrogen) The following secondary antibodies were used, goat anti-mouse Alexa488 /goat anti-rabbit Alexa568 /goat anti-rat Alexa647 /goat anti-rabbit Alexa647 /goat anti-rabbit Alexa480, 1:1000 (DSHB). Embryos were washed thrice with 0.3% PBS-T, and DAPI/Hoechst (1:500) was added in the penultimate wash. Embryos were mounted in 70% glycerol and observed under a Leica sp8 confocal microscope with a 20 X objective or Zeiss anisotropy confocal microscope under 63 x objective.

Embryos at the appropriate stage were washed and dechorionated as described previously. A 2-well Nunc Lab-Tek II Chamber Slide System (Thermo Fisher Scientific) was affixed with a 3 M Scotch double-sided tape, and dechorionated embryos with intact vitelline membranes were mounted on the tape under a dissecting microscope. Halocarbon oil 200 was used to cover the embryos to prevent dehydration during imaging. Time-lapse bright field imaging of embryos was performed on an inverted LSM confocal system (Zeiss multiphoton 710) at 20 X for ~8 hr. Embryo images were acquired at 5 min intervals.

Western blots were quantified using ImageJ. Each protein was normalized to the tubulin loading control, and the highest signal for each was set to 1 and all other tubulin band intensities were normalized against it. For MZT experiments, western blots were used to determine protein degradation dynamics. 3–5 biological replicates, representing 3–5 independent embryo collection were assayed. Band intensities were quantified using ImageJ and normalized to α-tubulin as loading control. For each replicate intensities were normalized to the first time point (0–1 hr). Significance thresholds are presented in the figures and figure captions.

All quantification of Smaug stained images were done on ImageJ. ROIs were made for a single pole cell through the z-stacks, thereby only analysing one pole cell at a time. For each embryo five such pole cells were measured. To quantify intensity, mean intensity was measured for each ROI per stack and added to get the final intensity. Each data point on the graph represents an average of all five cell intensities per embryo. For Smaug punctae measurement, a combined ROI for each cell was taken and duplicated. Each of the stack was then analysed using 3D objects counter, where thresholding for all images was kept within the range of 80–90 with a minimum size filter kept at zero. The number of objects counted was taken as the total number of Smaug punctae per cell. Five such pole cells were analysed per embryo and the average was plotted as a single data point on the graph. For quantification of poleplasm volume in the embryos, embryos were stained with Oskar. An ROI (kept same for all images) encompassing the posterior part of the embryo was drawn. Gaussian blur 3D (X sigma = 2, Y sigma = 2 and Z sigma = 2) was applied to each image. The volume of pole plasm was analysed by using 3D objects counter after appropriate thresholding.