The sea urchin provides a relatively simple and tractable system for analyzing the early stages of embryo development. Here, we use the sea urchin species, Paracentrotus lividus, to investigate the role of Alix in key stages of embryogenesis, namely the egg fertilization and the first cleavage division. Alix is a multifunctional protein involved in different cellular processes including endocytic membrane trafficking, filamentous (F)-actin remodeling, and cytokinesis. Alix homologues have been identified in different metazoans; in these organisms, Alix is involved in oogenesis and in determination/differentiation events during embryo development. Herein, we describe the identification of the sea urchin homologue of Alix, PlAlix. The deduced amino acid sequence shows that Alix is highly conserved in sea urchins. Accordingly, we detect the PlAlix protein cross-reacting with monoclonal Alix antibodies in extracts from P. lividus, at different developmental stages. Focusing on the role of PlAlix during early embryogenesis we found that PlAlix is a maternal protein that is expressed at increasingly higher levels from fertilization to the 2-cell stage embryo. In sea urchin eggs, PlAlix localizes throughout the cytoplasm with a punctuated pattern and, soon after fertilization, accumulates in larger puncta in the cytosol, and in microvilli-like protrusions. Together our data show that PlAlix is structurally conserved from sea urchin to mammals and may open new lines of inquiry into the role of Alix during the early stages of embryo development.
Cytoskeletal rearrangements, modulation of receptor-mediated cell signaling, and apoptosis are fundamental cellular processes involved in the early stages of animal development. In mammals Alix/AIP1 (ALG-2 interacting protein x/1) has been implicated in these diverse cellular events during cellular differentiation (Odorizzi 2006; Bongiovanni et al. 2012); however, until now the function/s of Alix during embryo development has not been elucidated. Alix is an evolutionarily conserved protein that derives its name from one of its interacting partners, the Apoptosis-Linked Gene-2 (ALG-2), a Ca++-binding protein involved in apoptosis (Missotten et al. 1999; Vito et al. 1999). The multitasking properties of Alix rely on its tripartite protein structure that allows its interaction with a multitude of partners. This domain organization is well conserved among Alix homologues and includes the N-terminal Bro1 domain, a proline-rich C-terminal region (PRR), and a central sequence containing two coiled-coil motif (V-domain) (Fisher et al. 2007; Pires et al. 2009). By interacting with the Endosomal Sorting Complexes Required for Transport (ESCRT) I and III at the Bro1 and PRR domains, respectively, Alix synergistically coordinates the endocytosis and recycling of membrane receptors, viral budding, and cytokinesis (Katoh et al. 2003, 2004; Martin-Serrano et al. 2003; Strack et al. 2003; Schmidt et al. 2004; McCullough et al. 2008). Relevant to the current study are the findings that Alix also participates in cytoskeleton remodeling by binding with F-actin at the Bro1 and PRR domains and other components of the cytoskeleton at the N-terminal half of its V-domain (Cabezas et al. 2005; Pan et al. 2006).
An Alix related protein, named ALX-1, has been detected in Caenorhabditis elegans and participates in the internalization and degradation of the receptor LIN-12 during patterning of the vulvar precursor cells (VPCs) (Shaye & Greenwald 2005). Similarly to its mammalian counterpart, ALX-1 is also associated with receptor recycling and the biogenesis of the multivesicular endosomes (MVEs) in the intestine (Shi et al. 2007). In Drosophila melanogaster, the Alix homologue forms a complex with the E3 ubiquitin ligase Plenty of SH3s (POSH) and ALG-2, and regulates the Jun N-terminal kinase (JNK) signal transduction pathway during the development of the imaginal eye discs (Tsuda et al. 2006).
Here, we explore the role of Alix in early embryogenesis, particularly at the developmental stages immediately following fertilization. We used the Mediterranean sea urchin P. lividus to show that Alix is conserved during evolution and that the sea urchin Alix homologue, PlAlix, is expressed in unfertilized eggs and throughout early embryonic development. Furthermore, we show that the localization of PlAlix changes dramatically after fertilization, becoming concentrated in specific regions. These findings suggest that PlAlix may function during early developmental events, a role that the protein may also exert in mammals.
Materials and methods
Identification of PlAlix
A P. lividus cDNA clone showing similarity to HuAlix was identified in a blast search of the Paracentrotus lividus expressed sequence tag (EST) database. The corresponding clone was picked up and sequenced. The sequence data have been submitted to the European Molecular Biology Laboratory (EMBL) database under accession number HE646599.
Sequence analysis, homology modeling, and phylogenetic tree construction
Amino acid sequences were aligned by the CLUSTALW program (http://www.ebi.ac.uk/Tools/msa/clustalw2). Comparative protein structure model for the PlAlix sequence was generated using the MODELLER-9v4 software, based on the crystallographic structure of HuAlix Bro1 and V domains (PDB code 2OEV).
Phylogenetic trees were inferred by the neighbor-joining (NJ) method, using MEGA5 (Tamura et al. 2011). Reliability of branching patterns in the NJ tree were assessed by 1000 bootstrap resamplings of the alignment data (Nei & Kumar 2000). Phylogenetic tree was constructed using the sequences of Alix/Pcdc6IP homologues identified by BLASTP homology search. Analysis was conducted with proteins of 10 different taxa.
Egg and embryo culture
Paracentrotous lividus sea urchins were collected along the north-western coast of Sicily. Gametes were harvested and eggs fertilized and cultured as previously described (Giudice 1973). Eggs used in immunostaining experiments were fertilized in the presence of 2 mmol/L PABA (para-aminobenzoic acid) to remove fertilization membrane. Total and post-nuclear proteins from eggs or embryos were extracted as previously described (Bongiovanni et al. 2012).
Immunoprecipitation and western blotting
60′ postfertilization embryos were lysed at 4°C for 30 min, using Hypotonic Buffer: 10 mmol/L Tris–HCl, pH 7.8, 1 mmol/L CaCl2, 5 mmol/L KCl, protease inhibitor cocktail (Complete, EDTA-free; Roche), Phosphatase Inhibitor Cocktail 1 and 2 (Sigma-Aldrich). Embryos were then disrupted by Dounce homogenization (40 strokes), and cellular debris and nuclei were removed by 10 min centrifugation at 800 g. MG132 (5 μmol/L final concentration), N-Ethylmalemide (10 mmol/L final concentration), and 2 mmol/L CaCl2 (final concentrations) were added to 1.5 mg of post-nuclear proteins from embryos in a final volume of 1 mL. Lysates were pre-cleared by 1 h incubation at room temperature with 25 μL of protein A/G Plus agarose beads (Santa Cruz Biotechnology) and then spun at 1000 g. 1.5 μg of anti-Alix (clone 3A9; Santa Cruz Biotechnology) or anti-Alix (clone 1A12; Santa Cruz Biotechnology) were added to the supernatant and incubated over night at 4°C, a sample without any antibody was used as negative control. To immunoprecipitate Alix, protein A/G Plus agarose beads were added and incubated at room temperature for 2 h (Santa Cruz Biotechnology). The beads were washed five times with Hypotonic Buffer supplemented with 0.1% NP-40, and the bound proteins detached. Immunoprecipitated proteins and total protein extracts were loaded onto sodium dodecyl sulfate (SDS)polyacrylamide gel and subjected to electrophoresis at 20 mA constant current, for 1 h. Proteins were wet-transferred at 90 mA overnight onto polyvinylidene difluoride membranes (PVDF, Hybond-P; GE Healthcare). Immunoblotting was performed as described (Bongiovanni et al. 2012) using a purified mouse anti-AIP1/Alix (BD Biosciences), produced against mouse Alix (immunogen: aa. 375–580) diluted 1:250, or rabbit anti-polyUbiquitin (THERMO Scientific), diluted 1:250. Quantification of bands was carried out, below saturation levels, using Quantity One software (Bio-Rad, Hercules, CA, USA).
Total RNA was isolated from unfertilized P. lividus eggs (UE) and embryos at different stages of development (5′, 30′, 60′, 120′ postfertilization) using TRI Reagent (Ambion), according to the manufacturer's instructions.
First strand cDNAs were retro-transcribed with SuperScript III Reverse Transcriptase (Invitrogen) using 2.5 μg of RNA. Quantitative real-time polymerase chain reaction (qPCR) was carried out on the cDNA equivalent of 8 embryos (25 ng), using Power SYBR Green Master Mix (Applied Biosystems). PlAlix primers were used to obtain a 110 bp length amplicon:
AL1500for 5′ TACCAGACCATTCTCAACAAT 3′
AL1610rev 5′ TGCTATTTCCGCTTCGCTTTT 3′.
Mitochondrial cytochrome oxidase gene primers were used to obtain a 91 bp length amplicon as an endogenous reference to normalize data (Cavalieri et al. 2011): Cyt-ox For2 5′ TTGGGGTTAATCTAACATTCTTC 3′:
Cyt-ox Rev2 5′ AGGGTATAGGCATCTGGATAG 3′.
qPCR experiments were carried out in triplicate using 75 nmol/L primers and a StepOnePlus Real-Time PCR System (Applied Biosystems); data were analyzed using the Applied Biosystems StepOnePlus. Real-Time PCR Software v2.1 was used in the comparative Ct method (ΔΔCt). The PCR cycle conditions were: initial denaturation at 95°C for 10 min, 40 cycles of 95°C for 15 s and 60°C for 1 min, and a final melting curve step.
Eggs and embryos were fixed in 4% paraformaldehyde in PBS (phosphate-buffered saline), washed three times in PBS, and blocked in 5% goat serum in PBS. Samples were incubated overnight a 4°C with anti-Alix (clone 3A9; Santa Cruz) diluted 1:20, or rabbit polyclonal anti-Alix (Bongiovanni et al. 2012) diluted 1:20 in blocking solution, washed in goat serum/PBS and incubated with Alexa Fluor 488 anti-mouse IgG (Invitrogen) or Alexa Fluor 594 anti-rabbit IgG (Invitrogen). F-actin was detected after permeabilization using 10 mmol/L fluorescein isothiocyanate (FITC)-phalloidin at room temperature for 60 min. Microtubules were labeled using monoclonal anti-α-tubulin (Sigma), diluted 1:25, and Alexa Fluor 488 anti-mouse IgG (Invitrogen) as secondary antibodies. After washing in goat serum/PBS and PBS, samples were mounted with VECTASHIELD mounting medium (Vector LAB). Specimens were examined by FluoView FV1000 confocal microscope (Olympus).
Identification, bioinformatic, and phylogenetic analyses of PlAlix
To identify the P. lividus Alix homologue, we used the human Alix (HuAlix) sequence to screen a P. lividus EST database containing partial sequences of a cDNA collection from unfertilized eggs and embryos (Marine Genomics Europe, NoE, project funded by European commission, FP6). The sequencing and analysis of an isolated, full length cDNA clone, from unfertilized egg cDNA library, showed that the predictive PlAlix protein consists of 896 amino acid residues, with a calculated molecular weight of 98.42 kDa that is very close to HuAlix (868 aa. residues and a calculated molecular weight of 96.02 kDa). The putative PlAlix has 85.5% identity compared with the predicted Alix protein of the sea urchin species, Strongylocentrotus purpuratus, and 45.1% identity and 63.3% similarity to its human counterpart. The multialignment of Alix homologous proteins from several taxa is reported in supplementary Figure S1.
To analyze the structure-function of PlAlix, we generated a comparative protein structure model of PlAlix based on the crystallographic structure of HuAlix Bro1 and V domains (PDB ID: 2OEV) (Fisher et al. 2007), using the MODELLER-9v4 software (Eswar et al. 2007). As in mammalian Alix proteins, the tripartite domain organization also occurs in PlAlix: N-terminal Bro1 domain, central coiled-coil V domain, and C-terminal Proline Rich Domain (PRD). The elements of the sea urchin and human Alix secondary structure (beta sheets and alpha helices) have been preserved, with lesser differences in the central V domain (Fig. 1A). The two hydrophobic regions (Patch 1 and Patch 2) on the Bro1 surface (Kim et al. 2005; Fisher et al. 2007) are also conserved in the Bro1 domain of PlAlix (Fig. 1A). Specifically, the docking site for the ESCRT-III component CHMP4 (charged multivesicular body protein 4) is well conserved in Patch 1, as well as the interaction domain for the SH2 of Src, within the Patch 2 around the Tyr321 (Fig. 1A). We also noted the presence of a conserved motif considered a putative Kinase A phosphorylation site (Mattei et al. 2005) and of a conserved region within the “hinge region” of the PlAlix V domain (Pires et al. 2009). Within the PRD domain (Fig. 1B), the amino acid residues that are known to be important for interactions with the ESCRT-I protein TSG101 (tumor susceptibility gene 101 protein), ALG-2 and the two retroviral Gag proteins (P6 and P9) (Odorizzi 2006), are also well conserved between HuAlix and PlAlix.
Phylogenetic analysis was performed using Alix homologues from different species. Alix homologues were identified by BLASTP search and downloaded from the GenBank database. Amino acid sequences were aligned by the CLUSTALW program and complete alignments are shown in Figure S1. Due to the low degree of conservation, the phylogenetic tree was reconstructed excluding the highly variable C-terminal PRD domain. Analysis was first conducted either with the Bro1 or the central V domain (data not shown), and ultimately with the whole region containing both domains (Fig. 1C). The topology of the tree and the high bootstrap values confirm that Alix from Echinodermata and Chordata share a common ancestor and that D. melanogaster and C. elegans Alix diverged earlier. Moreover, if compared with the Protostomes Alix, sea urchins Alix is closer to Chordates both in terms of topology and evolutionary distance expressed as the number of amino acid differences per site.
Identification and temporal expression of PlAlix in P. lividus embryos
To identify the PlAlix protein, western blots were performed using protein extracts from fertilized eggs and from mouse C2C12 muscle cells, used as positive control (Bongiovanni et al. 2012); denatured proteins were immunoblotted using a commercially available mouse monoclonal anti-Alix antibody (BD Biosciences). As shown in Figure 2A, a band of the expected molecular weight (about 100 kDa) comigrates with the mouse Alix protein present in C2C12 protein extract. To confirm the identity of PlAlix, native proteins from post-nuclear P. lividus fertilized eggs were immunoprecipitated using two additional anti-Alix monoclonal antibodies, in the presence of the proteasome inhibitor, MG132, and the de-ubiquitinating enzyme inhibitor, N-Ethylmalemide. Immunoblotting of these immunoprecipitated proteins confirmed the presence of a band corresponding to the PlAlix antigen in the P. lividus protein extracts (Fig. 2B). Immunoprecipitation of PlAlix was more efficiently driven by monoclonal antibody clone 3A9 than 1A12, which were both conformation-sensitive monoclonal antibodies raised against the full length HuAlix (Pan et al. 2006, 2008). This difference in efficiency may depend on the differential affinity of the two antibodies to the native PlAlix. Other bands at higher molecular weights were also immunoprecipitated with the two anti-Alix antibodies and, in line with the preservation of at least one putative ubiquitination site in PlAlix; they probably are ubiquitinated forms of Alix (see asterisks in Fig. 2B). This assumption is corroborated by the detection of bands with the same molecular weight using an anti-polyubiquitin antibody in immunoblots of the anti-Alix immunoprecipitates (Fig. 2B, right panel). In addition, the relative ratios of PlAlix versus ubq-PlAlix forms differ significantly between 1A12- and 3A9-pulled-down samples; this difference can reflect the exposure of specific epitopes in ubiquitinated forms of PlAlix. It is worth noting that bands of similar size, positive for polyUbiquitin, were also detected in anti-Alix immunoprecipitates of C2C12 protein extracts, and that Alix undergoes efficient ubiquitination by three different ubiquitin ligases, including Ozz-E3 (Bongiovanni et al. 2012).
To analyze the expression pattern of PlAlix during the early stages of development, equal amounts of protein extract from unfertilized eggs and embryos at different times following fertilization (1, 5, 15, 30, 60 and 120 min) were immunoblotted using monoclonal anti-Alix; immuno-positive Alix bands were quantified and normalized against the Ponceau S loading control. As shown in a representative gel and in the quantitative analysis of PlAlix bands (Fig. 2C, D), a significant increase of PlAlix protein was observed from unfertilized eggs until 60–120 min after fertilization, indicating a regulated expression during these early stages of development. In these experiments, PlAlix bands are associated with lower molecular mass bands that may be cleavage products of PlAlix (Fig. 2C, gray arrowhead), in line with what has been reported for HuAlix (Pan et al. 2008). Furthermore, immunoblots of protein extract from embryos at later developmental stages (128-cell, blastula, and gastrula stage embryos) showed that PlAlix is present and its level remains unaltered during these later stages (Fig. 2E). To see if PlAlix expression was similarly regulated at the transcriptional level, we analyzed the relative abundance of the PlAlix transcript in unfertilized eggs (UE) and embryos up to 120′. Total RNA from each stage was retro-transcribed and used in qPCR for PlAlix transcripts, normalized with cytochrome oxidase (Cavalieri et al. 2011). Our results show that, in contrast to protein expression, PlAlix mRNA level decreased to <20% in 5′ post-fertilization embryos compared to the unfertilized eggs, and remained at low levels in all the examined developmental stages (Fig. 2F).
Subcellular localization of PlAlix during P. lividus development, from zygote to 2-cell stage embryo
To examine the subcellular distribution of PlAlix in the P. lividus embryo, we performed whole-mount immunofluorescence labeling of eggs and early embryos soon after fertilization (15 and 60 min). Unfertilized eggs displayed a punctuated staining of PlAlix throughout the cytosol (Fig. 3A, E). In early embryos PlAlix accumulated throughout the cytosol and localized in enlarged puncta near the nuclei, along the spindle microtubules (Fig. 3B, C, K, L, and M). Thus, PlAlix is localized in cellular structures, likely intracellular vesicles, of different sizes: the smallest are also present in unfertilized eggs, while others of larger size appear soon after fertilization and are more evident after the first division.
Close inspection of several specimens pointed out that a number of PlAlix-positive structures are localized close to the cortex or extended from the membrane surface (see arrowheads in Fig. 3E–G); the number of these sub-cellular compartments increased after fertilization (Fig. 3F, G versus E panel). In order to shed light on the identity of these Alix-positive regions, we investigated on Alix localization in relation to actin-containing structures. We observed that PlAlix is in close proximity with F-actin in several cellular protrusions which resemble microvilli (H–J). Furthermore, we observed that PlAlix staining increased in intensity and distance from the membrane surface after fertilization and until 60 min post-fertilization.
Cellular events such as membrane trafficking, endocytosis/exocytosis and the remodeling of actin filaments are fundamental processes during oogenesis, fertilization, and early embryogenesis. While the function of Alix has been established in all these phenomena, in differentiated mammalian cells (Odorizzi 2006; Bongiovanni et al. 2012), until now nothing it is known about its function/s during the early stages of embryo development. In this study, we used the Mediterranean sea urchin species, P. lividus, as the animal system to investigate the role of Alix in key stages of embryogenesis, namely the egg fertilization and the first cleavage division. We provide evidence for the presence of Alix in sea urchin, by comparing human Alix sequences against EST databases of the Paracentrotus lividus species. The sequencing and the analysis of the isolated cDNA clone confirmed this evidence. Immunoblotting, immunoprecipitation, homology modeling and phylogenetic analyses also showed that the protein PlAlix is well conserved during evolution. In this context, it is worth noting that a region of similarity was indicated by the three dimensional structure of the highly hydrophobic pocket within the “hinge region” of the PlAlix V-domain (Pires et al. 2009); this region, around Lys642 in PlAlix, is well preserved. This finding suggests that PlAlix might have preserved an activation/localization mechanism as the mammalian Alix. Indeed, it has been recently suggested that opening of the ‘hinge region’ of Alix, by posttranslational modifications and/or ligand binding, would expose different binding sites and consequently modify the function and subcellular localization of Alix (Fisher et al. 2007; Zhai et al. 2011). In line with this concept, we have recently found that the substrate-recognition module of the Ozz-E3 ubiquitin ligase, Ozz, binds to and ubiquitinates Alix, affecting the conformation of Alix and controlling its local concentration at sites where cytoskeleton remodeling is taking place (Bongiovanni et al. 2012). In this regard, it is noteworthy that the preserved Lys642 of PlAlix is included in the “hinge region” of the V-domain (Fig. 1A). According to the predictor of ubiquitination sites, UbPred (Radivojac et al. 2010), this lysine, together with six others (Lys146, Lys411, Lys422, Lys468, Lys471, Lys709), is predicted to be an ubiquitination site with more than “medium confidence”. Furthermore, the finding that the putative ubiquitinated Lys642, together with Lys422 and Lys709, is well-preserved among different species makes it very attractive as a possible regulation site for the conformation and function of Alix.
Next, we examined the expression pattern of PlAlix during the early stages of development. The gradual increase of the PlAlix protein level immediately after fertilization and until the 2-cell stage points to a possible active role of Alix in early embryo development. In contrast, we observed a severe decrease in PlAlix mRNA abundance during these early stages of development, probably as a result of the general maternal-to-zygotic transition and the loss of maternal mRNAs (Aguirre-Armenta et al. 2011). Taken together, these results are in agreement with the elevated efficiency of protein synthesis after fertilization (Metafora et al. 1971; Raff et al. 1981). Furthermore, protein stability could be another factor that would contribute to the overall abundance of PlAlix in early developmental stages.
The immunofluorescence experiments revealed that the cytosolic staining of PlAlix was noticeably remodeled if we compared eggs and early embryos; it gradually switched from diffuse and punctuated to mostly punctuated, with an increase in the number of larger spots soon after fertilization. These PlAlix-positive large spots appeared 15 min after fertilization and become more evident at the first embryonic cleavage; they were distributed near the two nuclei, specifically in the central region along the spindle microtubules. This localization is in agreement with what is already known about the distribution of Alix in vesicles of endocytic trafficking in different species, from Dictyostelium to mammalians (Mattei et al. 2005; Shaye & Greenwald 2005; Montagnac et al. 2008). The subcellular distribution of PlAlix is also compatible with the presence of a large number of small, maternally stored granules and vesicles randomly distributed in sea urchin unfertilized eggs (Ramos et al. 2010). Furthermore, the redistribution of cytoplasmic components upon fertilization is a common phenomenon in a variety of animal eggs and also in sea urchins, in which a whole spectrum of acidic organelles and intracellular vesicles are present (Morgan 2011). These cellular compartments are engaged in a highly organized series of vesicular transport mechanisms that are activated at different times (Matese et al. 1997; Conner & Wessel 1998). Among these vesicles, small endosomes are already present in the unfertilized egg, even though endocytosis is slow and possibly constitutive (Covian-Nares et al. 2008). Following the massive exocytosis at fertilization, the regulated compensatory endocytosis pathway takes place; it is promptly able to retrieve the excess cortical granule membranes and generate large endosomes, up to 3 μm in diameter (Whalley et al. 1995). Within this context, the localization of PlAlix in presumptive small and large endocytic vesicles suggests that its presence is in both constitutive endosomes (small vesicles) and compensatory endosomes (large vesicles). The presence in PlAlix of a well-conserved domain needed to bind with the ESCRT-III protein, CHMP4, is further evidence that Alix is likely to play an active role in endosomal functioning during development.
We and others have reported that Alix regulates the F-actin cytoskeleton and membrane dynamics during both endocytosis and the formation of cellular protrusions (Odorizzi 2006; Bongiovanni et al. 2012). The presence of discrete microvilli-like regions positive for PlAlix in close proximity to the membrane surface or protruding from it, suggests a possible role of PlAlix in the biogenesis and dynamics of egg/embryo microvilli. Microvilli are thousands of finger-like protrusions, which contain an internal ultrastructure of actin filament bundles, and cover the egg surface (Burgess & Schroeder 1977; Schroeder 1978). We observed that PlAlix staining increased in intensity and distance from the membrane surface after fertilization and until 60 min postfertilization. This is in accordance to what was previously reported, that microvilli covering the surface of unfertilized eggs increase in extent and diameter between 30 and 90 min after fertilization (Schroeder 1978). A further insight about the identity of the PlAlix-positive protrusions is the finding that PlAlix and F-actin staining are colocalized or juxtaposed in some regions of the plasma membrane or protruding from it in the extracellular space. This colocalization is consistent with the ability of Alix to interact with both F-actin and membrane components (e.g., endophilin) (Odorizzi 2006). However, while the function of Alix has been established in differentiated mammalian cells and it has only been inferred for very late stages of embryo development (Shaye & Greenwald 2005; Odorizzi 2006; Tsuda et al. 2006), our data indicate, for the first time, a role of Alix during early embryo development, namely from fertilization to the 2-cell stage embryo. Indeed, based on our findings and on what is already known about mammalian Alix, we infer that PlAlix may play a function as a regulator of early developmental events, such as the biogenesis of microvilli and the endocytic pathway, a role that the protein may also exert during mammalian early embryogenesis. Thus, we believe that this work may provide new insights into the fundamental mechanisms involved in early embryo development, in particular from fertilization to the 2-cell stage.
We thank Dr Maria Ina Arnone (SZN, Naples) for keeping the cDNA libraries and providing the PlAlix containing clone. Dr Yvan Campos (St. Jude C.R.H., Memphis, USA) for helpful discussion and Dr Xiaohui Qiu (St. Jude C.R.H., Memphis, USA) for the purification of the rabbit polyclonal anti-Alix antibody. We thank Dr Richard Burket for editing the manuscript. We also thank Dr Rosalia La Tona for her contribution to the qPCR experiments. This work has been partially supported by a grant from the Italian Ministry of Economy and Finance to the CNR for the Project FaReBio di Qualità.