Evidence that fertilization activates starfish eggs by sequential activation of a Src-like kinase and phospholipase cgamma.

Recent evidence has indicated a requirement for a Src family kinase in initiating Ca(2+) release at fertilization in starfish eggs (Giusti, A. F., Carroll, D. J., Abassi, Y. A., Terasaki, M., Foltz, K. R., and Jaffe, L. A. (1999) J. Biol. Chem. 274, 29318-29322). We now show that injection of Src protein into starfish eggs initiates Ca(2+) release and DNA synthesis, as occur at fertilization. These responses depend on the phosphorylation state of the Src protein; only the kinase active form is effective. Like Ca(2+) release at fertilization, the Ca(2+) release in response to Src protein injection is inhibited by prior injection of the SH2 domains of phospholipase Cgamma. These findings support the conclusion that in starfish, sperm-egg interaction causes egg activation by sequential activation of a Src-like kinase and phospholipase Cgamma. Injection of the SH2 domain of Src, which inhibits Ca(2+) release at fertilization, does not inhibit Ca(2+) release caused by Src protein injection. This indicates that the requirement for a Src SH2 domain interaction is upstream of Src activation in the pathway leading to Ca(2+) release at fertilization.

At fertilization, signals at the site of sperm-egg interaction cause a rise in cytosolic Ca 2ϩ (1,2). This opens ion channels and stimulates exocytosis of cortical granules, resulting in blocks to polyspermy, and also stimulates the resumption of the cell cycle (3)(4)(5). The Ca 2ϩ rise results, at least in large part, from Ca 2ϩ release from the endoplasmic reticulum, mediated by inositol 1,4,5-trisphosphate (IP 3 ) 1 (6 -10). Much recent work on fertilization has focused on the signal transduction pathways that lead to IP 3 production.
The phospholipase C family of enzymes produces IP 3 and diacylglycerol from the membrane lipid phosphatidylinositol 4,5-bisphosphate (11). In echinoderm eggs, it is the ␥ isoform of phospholipase C (PLC␥) that functions at fertilization. PLC␥ enzyme activity increases by 30 s post-fertilization in sea urchin eggs (12), and inhibition of PLC␥ activation inhibits Ca 2ϩ release at fertilization in both sea urchin and starfish eggs (13)(14)(15). In these experiments, PLC␥ activity was inhibited by injection of eggs with excess Src homology 2 (SH2) domains of PLC␥. SH2 domains are found in many signaling proteins, and provide a site for specific interaction of a particular protein with a particular phosphorylated tyrosine on another protein (16). Excess SH2 domains, introduced into cells by microinjection, function as specific dominant negative inhibitors of such interactions.
PLC␥ can be activated by phosphorylation of a regulatory tyrosine, although other factors may also be significant (17)(18)(19)(20). In sea urchin eggs, attempts to determine if PLC␥ is tyrosine phosphorylated at fertilization have been inconclusive, since the phosphotyrosine in PLC␥ immunoprecipitates was barely detectable either before or after fertilization (12,21). As discussed by these authors, a local increase at the site of spermegg interaction might have been too small to detect by the methods used. Nevertheless, tyrosine kinase activity increases within 15 s after fertilization (22), and the tyrosine kinase inhibitor genistein delays Ca 2ϩ release at fertilization (23). One group of tyrosine kinases that participates, directly or indirectly, in activation of PLC␥ is the Src family (20, 24 -26), and in vitro experiments with starfish eggs have shown a fertilization-dependent association of a Src-like kinase with the SH2 domains of PLC␥ (27). Further evidence that a Src family kinase functions to activate PLC␥ at fertilization comes from findings that in both starfish and sea urchin eggs, injection of excess SH2 domains of Src family kinases inhibits Ca 2ϩ release at fertilization (28,29). In addition, in sea urchin eggs, the Src family kinase inhibitor PP1 delays Ca 2ϩ release at fertilization, and the activity of a Src-like kinase increases by 30 s postinsemination (29).
These findings indicate that a Src-like kinase may, directly or through intermediate molecules, activate PLC␥ at fertilization, leading to Ca 2ϩ release and egg activation. In this report, we examine four questions related to this model. 1) Is the kinase activity of the Src-like protein required for Ca 2ϩ release? 2) Does a Src-like protein initiate DNA synthesis as well as Ca 2ϩ release? 3) Does the Src-like protein act upstream of PLC␥? 4) Do the SH2 domains of the Src-like protein interact with an upstream or downstream component of the pathway?
We approached these questions by injecting starfish eggs with human cSrc protein that was produced in insect cells. The activity of Src family kinases is regulated by phosphorylation of two tyrosines: phosphorylation of an internal tyrosine (Tyr-419 in human cSrc) is stimulatory, while phosphorylation of the COOH-terminal tyrosine (Tyr-530) is inhibitory (30). A third tyrosine phosphorylation site has also been identified, but its functional significance is not known (31). We injected starfish eggs with Src protein that was either completely unphosphorylated, or phosphorylated on Tyr-530 only. Unphosphorylated Src protein will, in the presence of ATP in the cytoplasm, rapidly autophosphorylate on Tyr-419, resulting in a fully ac-tive protein (Ref. 32 and see "Discussion"). Phosphorylation of Tyr-530 locks Src in a closed conformation in which intramolecular interactions among the SH3, SH2, and kinase domains down-regulate the kinase activity (33)(34)(35)(36).
We found that injection of starfish eggs with unphosphorylated Src protein causes Ca 2ϩ release and DNA synthesis. We then injected eggs with the SH2 domains of PLC␥, followed by unphosphorylated Src protein, to investigate whether Src releases Ca 2ϩ by way of PLC␥. We also injected eggs with the SH2 domains of Src, followed by unphosphorylated Src protein, to determine whether, in the pathway leading to PLC␥ activation at fertilization, the requirement for the Src SH2 domain is upstream or downstream of the activation of the Src-like kinase.

EXPERIMENTAL PROCEDURES
Recombinant Proteins-An NH 2 -terminal deletion mutant of human cSrc (N-85-srcTK, amino acids 86 -536) was expressed in baculovirusinfected Sf9 cells (31). The Src fraction that was phosphorylated on Tyr-530, and not on the other two tyrosine phosphorylation sites, was purified as described previously (33). Unphosphorylated Src was purified with a combination of ATP-Sepharose, phosphotyrosine, and anionexchange chromatography. The phosphorylation state of the protein fractions was confirmed by matrix-assisted laser desorption/ionization mass spectrometry after trypsin digestion, two-dimensional phosphopeptide mapping, kinase assay, and immunoblotting using a Tyr-530 phosphospecific Src antibody (44-662, 0.25 g/ml; BIOSOURCE International, Camarillo, CA). A non-phosphospecific Src antibody (SC-18, 0.4 g/ml; Santa Cruz Biotechnology, Inc., Santa Cruz, CA) was used as a control. The blots were developed using a horseradish peroxidaseconjugated antibody against rabbit IgG (SC-2030, Santa Cruz Biotechnology), and ECL reagents (Amersham Pharmacia Biotech). Src protein solutions used for injection contained 7.5-8.0 mg/ml protein in 20 mM Hepes (pH 7.5), 0.1 M NaCl, and 5 mM dithiothreitol. Protein concentrations were determined using a BCA assay (Pierce Chemical Co., Rockford, IL) with bovine serum albumin as the standard.
SH2 domain glutathione S-transferase fusion proteins were made in bacteria and purified as described previously (13,28). SH2 domain protein solutions for injection contained 33 mg/ml protein and 330 M calcium green dextran in 137 mM NaCl, 2.7 mM KCl, 8.1 mM Na 2 HPO 4 , and 1.5 mM KH 2 PO 4 (pH 7.2).
Microinjection-Starfish (Asterina miniata) were obtained from Marinus, Inc. (Long Beach, CA) and from Will Borgeson (Bodega Marine Lab, Bodega Bay, CA). Oocytes were collected as described previously (13). Quantitative microinjection was performed using mercury-filled micropipettes (37,38). Calcium green dextran, Oregon green dUTP, and SH2 domain proteins were injected into immature oocytes; 1 M 1-methyladenine (Sigma) was then applied to cause oocyte maturation. In one set of experiments, we confirmed that PLC␥ SH2 domains have the same effect on Ca 2ϩ release, whether they are injected before or after oocyte maturation (see also Ref. 13). Src proteins were injected into mature eggs at first meiotic metaphase, 40 -160 min after the initial injection. Injection volumes were 1-5% of the egg volume (3100 picoliters), and concentrations given in the text indicate the final values in the egg cytoplasm. IP 3 (Calbiochem, San Diego, CA) was used as a 10 M stock, and 5% of the egg volume was injected. All experiments were performed at 16 -18°C, with the eggs in natural sea water.
DNA Synthesis Measurements-DNA synthesis was detected in eggs preinjected with 1-2 M Oregon green 488-5-dUTP (Molecular Probes) as described previously (14). After injection of Src protein, eggs were cultured for 2.5-5 h at 16 -18°C. DNA synthesis was detected by the observation of fluorescent chromatin, using a confocal microscope (Bio-Rad). The image shown in Fig. 4 was obtained with a ϫ40, 1.3 N.A. objective (488 nm, low laser power, 10% neutral density filter, zoom 2.5). The presence or absence of DNA synthesis was confirmed by photobleaching to remove unincorporated Oregon green dUTP from the cytoplasm (14). If fluorescent chromatin was not visible, the nucleus was located by coinjection of 70-kDa Texas Red dextran (Molecular Probes), which is excluded from the nucleus (13). The photobleaching area (78 ϫ 43 m) was positioned away from the nucleus or chromatin mass, and the egg was exposed to 7 sets of 10 scans (ϫ20, 0.5 N.A. objective, 488 nm, full laser power, zoom 8.5, slow scan setting). After photobleaching, eggs that had synthesized DNA retained fluorescence in the chromatin, while eggs that had not synthesized DNA showed no nuclear fluorescence. The figure was prepared using Photoshop (Adobe).

Injection of Starfish Eggs with Kinase Active Src Protein
Causes Ca 2ϩ Release-Human c-Src protein (amino acid residues 86 -536) was produced in insect cells, and two major phosphorylation forms were purified: unphosphorylated (P Ϫ Src) and monophosphorylated at Tyr-530 (PY530Src). Both phosphorylation forms contain the catalytic, SH2, and SH3 domains, but lack the NH 2 -terminal hydrophobic domain by which Src associates with membranes ( Fig. 1). In the presence of the ATP in the cytoplasm, P Ϫ Src will rapidly autophosphorylate on Tyr-419 and acquire high enzymatic activity (see "Discussion").
Injection of P Ϫ Src into starfish eggs (370 g/ml ϭ 7.2 M final concentration) caused Ca 2ϩ release in all eggs tested (n ϭ 31), beginning ϳ1.6 min after injection (Fig. 2B, Table I). Ca 2ϩ was detected using calcium green dextran; the fluorescence reached a peak that was 74 Ϯ 18% greater than the baseline level (S.D., n ϭ 31). This Ca 2ϩ increase was somewhat smaller than that seen at fertilization ( Fig. 2A), where the peak fluorescence increase was 102 Ϯ 15% (n ϭ 5), using the same optical measurement conditions. The Ca 2ϩ rise caused by Src protein injection usually lasted for several minutes, although the duration was generally somewhat shorter than at fertilization (compare Fig. 2, A and B). Src protein injection also caused partial or complete elevation of the fertilization envelope, indicating the occurrence of partial or complete cortical granule exocytosis. Ca 2ϩ release in response to injection of Src protein was concentration dependent, and occurred in only 50% of eggs injected with P Ϫ Src at a final concentration of 220 g/ml (4.3 M). Ca 2ϩ release was not detected when the Src protein concentration was reduced to 75 g/ml (1.5 M) (Table I).
Confocal imaging of calcium green dextran fluorescence following Src injection (370 g/ml) showed that the Ca 2ϩ rise occurred in a wave that started after the characteristic delay described above, from a site near the plasma membrane toward which the solution had been expelled from the micropipette (Fig. 3) (n ϭ 6). The Ca 2ϩ wave closely resembled that seen at fertilization (see Refs. 13 and 28). It differed from the pattern of Ca 2ϩ release seen when eggs were injected with IP 3 , where the Ca 2ϩ release began immediately, and from the site of injection rather than from the egg surface ( Fig. 3) (n ϭ 4).
To examine whether the kinase activity of the Src protein was required for the Ca 2ϩ release in response to injection of the protein, we purified, from the mixture of Src forms produced by the insect cells, Src protein that was monophosphorylated on the inhibitory tyrosine at the COOH terminus (PY530Src) (Fig.  1, B and C). In this phosphorylation state, the Src tyrosine kinase activity is down-regulated, even in the presence of ATP (30,32,39). Upon injection into starfish eggs (400 g/ml ϭ 7.8 M), PY530Src caused little or no Ca 2ϩ release; any calcium green dextran fluorescence increase that occurred was always less than 10% of the baseline fluorescence (Fig. 2C, Table I).
These results indicate that kinase activity is required for Ca 2ϩ release in response to Src injection.
Src Protein Injection Causes DNA Synthesis-The resumption of the cell cycle is a common feature of fertilization in all species; in the starfish A. miniata, this is marked by the occurrence of DNA synthesis at about 2 h after fertilization (40,41). We examined whether injection of starfish eggs with P Ϫ Src protein caused DNA synthesis.
DNA synthesis was detected by preinjecting the eggs with a fluorescent nucleotide analog, Oregon green dUTP (see Ref. 14, and "Experimental Procedures"). 2.5-5 h after injection of active Src protein (P Ϫ Src), we examined the eggs using confocal microscopy. In 15 of 18 eggs, a condensed cluster of Oregon green dUTP-labeled chromatin was visible (Fig. 4). In 4 of these eggs, we confirmed that DNA synthesis had occurred by photobleaching to remove unincorporated Oregon green dUTP. Photobleaching in a region of the egg cytoplasm away from the chromatin did not remove the chromatin fluorescence; this showed that the fluorescent nucleotides in the chromatin region were no longer diffusible, indicating that they had been incorporated into DNA (see Ref. 14, and "Experimental Procedures"). In contrast, a parallel set of experiments showed that eggs injected with catalytically inhibited Src protein (PY530Src) did not undergo DNA synthesis (n ϭ 3). In the PY530Src injected eggs, no Oregon green dUTP fluorescence remained in the egg cytoplasm after photobleaching. These  a Significantly different from that for eggs injected with P Ϫ Src alone, or with SHP2SH2 followed by P Ϫ Src (p Ͻ 0.0002). results showed that elevating Src kinase activity in starfish eggs stimulates DNA synthesis, as occurs at fertilization. Although some eggs subsequently showed multiple nuclei or irregular cleavage, no further development was observed. The failure of the Src-injected eggs to undergo normal cell division and development might be due to the absence of the sperm centriole that normally provides the mitosis organizing center in the fertilized egg (42).
Src Acts Upstream of PLC␥-To examine whether the Ca 2ϩ release in response to injection of P Ϫ Src occurred by the same pathway as at fertilization, we investigated whether the P Ϫ Src response was inhibited by preinjection of the SH2 domains of PLC␥. As described previously (13), injection of PLC␥ SH2 domains (1 mg/ml) delays and reduces Ca 2ϩ release at fertilization (Fig. 5A). Injection of PLC␥ SH2 domains (1 mg/ml) also had an inhibitory effect on Ca 2ϩ release following injection of P Ϫ Src (Fig. 5B, Table I). Two of the 10 eggs tested showed no Ca 2ϩ release in response to P Ϫ Src injection. Eight of the 10 eggs eventually released Ca 2ϩ , but the delay between injection of the P Ϫ Src and Ca 2ϩ release was significantly longer than in eggs containing the SH2 domains of a control protein, the phosphatase SHP2 (Fig. 5C, Table I). Tests of PLC␥ and SHP2 SH2 domains were carried out alternately in each set of experiments. Although the peak amplitude of the calcium green fluorescence in P Ϫ Src injected eggs containing PLC␥ SH2 domains was not significantly different from that in eggs containing control SH2 domains, the Ca 2ϩ elevation in the PLC␥ SH2-injected eggs was usually shorter in duration and often consisted of several brief peaks instead of the sustained rise observed in the control-injected eggs (compare Fig. 5, B and C). The increase in the delay between P Ϫ Src injection and Ca 2ϩ release, caused by PLC␥ SH2 domains, indicates that both fertilization and P Ϫ Src initiate Ca 2ϩ release by way of PLC␥.
The SH2 Domain of Src Interacts with an Upstream Regulator in the Pathway Leading to Ca 2ϩ Release-Injection of starfish eggs with the SH2 domain of Src also delays and inhibits Ca 2ϩ release at fertilization (Ref. 28; Fig. 6A). This could result from an inhibition of the interaction of a Src-like kinase either with an upstream regulator of Src activation, or with a downstream target of the activated kinase. To examine these alternatives, we investigated whether injecting eggs with Src SH2 domains inhibited Ca 2ϩ release in response to subsequent injection of P Ϫ Src. We found that Src SH2 domains did not prevent the Ca 2ϩ rise in response to P Ϫ Src, and did not significantly increase the delay between P Ϫ Src injection and the Ca 2ϩ rise (Fig. 6B Table I), or reduce the amplitude of the Ca 2ϩ rise relative to that in eggs preinjected with SH2 domains of a control protein, the tyrosine kinase Abl. These results support The second image pair shows the moment of injection as indicated by the oil droplet at the tip of the micropipette. In the P Ϫ Src sequence, a small Ca 2ϩ transient at the injection site was followed by a rapid return to baseline. No further Ca 2ϩ increase was observed until 70 s postinjection, when a wave of Ca 2ϩ release began from the right side of the egg. In the IP 3 sequence, Ca 2ϩ release began at the injection site, immediately after injection, and spread out to the plasma membrane by 2 s post-injection. Scale bar ϭ 100 m. Quicktime movies showing these sequences are available online. Each movie is composed of confocal images taken at 2 frames/s, and played back at 10 frames/s (5 ϫ real time). the conclusion that in the pathway leading to PLC␥ activation and Ca 2ϩ release at fertilization, the requirement for a Src SH2 domain interaction is upstream of the activation of the endogenous Src-like kinase.

DISCUSSION
Src Tyrosine Phosphorylation and Kinase Activity-By injecting starfish eggs with the tyrosine kinase Src, we have demonstrated that Src kinase activity is sufficient to initiate Ca 2ϩ release quite similar to that at fertilization, and to initiate DNA synthesis as occurs at fertilization. Two forms of the Src protein were used, P Ϫ Src, in which none of the tyrosines are phosphorylated, and PY530Src, which is phosphorylated on Tyr-530 only. Upon exposure to ATP in the egg cytoplasm, Ͼ80% of P Ϫ Src is expected to rapidly autophosphorylate on Tyr-419, leading to activation of enzymatic activity (32). In vitro at 25°C, in the presence of 1 mM ATP (comparable to that in the egg cytoplasm; Ref. 43), Tyr-419 phosphorylation occurs within Ͻ2 min (32). Thus the time required for autophosphorylation could account for a part of the ϳ1.6 min delay that we observe between injection of the Src protein and the Ca 2ϩ rise. Diffusion of the Src protein in the egg cytoplasm, and the time required for steps leading to PLC␥ activation and IP 3 -induced Ca 2ϩ release, could also contribute to the delay. The Ca 2ϩ rise in response to P Ϫ Src injection, while similar to that at fertilization, is not identical, being somewhat smaller in amplitude and duration. This could reflect the fact that in the presence of ATP, Src protein gradually autophosphorylates on Tyr-530 and autodephosphorylates on Tyr-419, such that the maximally active form of the protein is transient (32). Furthermore, our injection conditions do not precisely mimic the conditions under which a Src family kinase may be activated at fertilization; for example, in the continuing presence of an activator of Src, the PY419 state might be maintained for a longer time. Decreased membrane binding of the injected Src protein, due to its lack of a myristoylation site at the amino terminus, could be another significant factor.
PY530Src is also expected to undergo some autophosphorylation on Tyr-419 when exposed to ATP in the egg cytoplasm, 2 but the activity of Src that is phosphorylated on both Tyr-419 and Tyr-530 is Յ20% of that for Src phosphorylated on PY419 only (32). Therefore, injection of PY530Src should introduce much less kinase activity in the egg cytoplasm compared with injection of P Ϫ Src. Correspondingly, PY530Src did not cause Ca 2ϩ release.
In previous studies, Src family kinases have been introduced into cells by viral infection, transfection of DNA, and injection of RNA. These studies have shown that overexpression of constitutively active Src mutants causes various downstream cellular responses, including unregulated cell division and cytoskeletal rearrangements (e.g. Refs. 44 and 45). Here we introduced Src into a cell directly as a purified recombinant protein, allowing us to study a rapid response (Ca 2ϩ release) to a step increase in the amount of Src protein in the cell, and to test the differential effects of two distinct phosphorylated forms of Src. Only the kinase active form of Src results in Ca 2ϩ release and initiation of DNA synthesis.
Signaling Pathways at Fertilization-As summarized in the Introduction, recent evidence indicates requirements for both PLC␥ and a Src family kinase in the signaling pathway leading to Ca 2ϩ release at fertilization. The findings reported in this paper establish that the Src-like kinase acts upstream of PLC␥, since injection of PLC␥ SH2 domains prevents or delays Ca 2ϩ release in response to injection of active Src protein. What intermediate molecules may function in this pathway are unknown, but studies of immune cells and platelets suggest that intermediate kinases such as Syk or ZAP-70, and/or linker proteins such as LAT or SLP-76, may be involved (25,26,46). Injection of starfish eggs with SH2 domains of mammalian Syk and ZAP-70 does not inhibit Ca 2ϩ release at fertilization (28), but there may be different intermediate kinases in the starfish egg. A kinase cascade, if it included a positive feedback loop, could serve to amplify a local signal at the site of sperm-egg interaction (see Ref. 47).
PLC␥ activation at fertilization leads to IP 3 production and Ca 2ϩ release from the endoplasmic reticulum (see Introduction). Consequences of the Ca 2ϩ rise include exocytosis of cortical granules, which establishes a block to polyspermy (3), and inactivation of mitogen-activated protein kinase, which leads to the initiation of DNA synthesis (5,14,41,48). PLC␥ activation also results in production of diacylglycerol, which may stimulate other egg activation events (see 49).
This model for echinoderm egg activation at fertilization applies to vertebrate eggs in some but not all aspects. At fertilization, vertebrate eggs also produce IP 3 and release Ca 2ϩ , leading to cortical granule exocytosis and resumption of the cell cycle (4,9,50,51). However, Ca 2ϩ release at fertilization in frog and mouse eggs is not inhibited by excess PLC␥ SH2 domains, indicating that if PLC␥ is activated, it is not by an SH2 domain-dependent mechanism (52,53). Nevertheless, in frog eggs, a Src-like kinase becomes tyrosine phosphorylated within 1 min of insemination (54), and tyrosine kinase inhibitors inhibit Ca 2ϩ release and other activation events at fertilization (55)(56)(57). Likewise, both PLC and tyrosine kinase inhibitors show some inhibitory effects on Ca 2ϩ release at fertilization in mouse eggs (58). These findings indicate that tyrosine kinases function in vertebrate fertilization, but the connection between these kinases and IP 3 production is not understood.
Initiation of Src Family Kinase Activation at Fertilization-Injection of starfish eggs with excess SH2 domains of Src, which inhibits Ca 2ϩ release at fertilization, does not inhibit Ca 2ϩ release caused by Src protein injection. This indicates that the Src SH2 domain functions upstream of the activation of the Src family kinase in the pathway leading to Ca 2ϩ release at fertilization. Therefore, whatever is directly upstream of the 2 W. Xu, unpublished results. Src family kinase should have a binding site for the Src SH2 domain.
Activation of Src family kinases in cells is hypothesized to occur by at least four different means (30, 33, 35, 59 -61). 1) Tyrosine 419 might be phosphorylated, causing a conformational change that activates the kinase. 2) Dephosphorylation of the COOH-terminal tyrosine (Tyr-530) could release this tyrosine from its intramolecular association with Src's SH2 domain, and thus disrupt the inhibitory closed conformation of Src and result in an active tyrosine kinase. 3) Interaction of Src's SH2 domain with a phosphorylated tyrosine on another protein could outcompete the binding of the SH2 domain to the COOH-terminal tyrosine, and thus disrupt the closed conformation. 4) Interaction of Src's SH3 domain with a high affinity proline-rich ligand might also open up Src's protein structure. Our findings support the third possibility, since this model would account for the inhibition of Ca 2ϩ release at fertilization by excess SH2 domains, and the lack of effect of excess SH3 domains (28).
Proteins that activate Src by binding to Src's SH2 domain include the platelet-derived growth factor receptor (see Ref. 59), antigen receptors, by way of their "immune receptor tyrosine activation motifs" (62), and the focal adhesion kinase FAK (63,64). Possibly sperm-egg interaction results in the tyrosine phosphorylation of such a molecule in the egg, allowing it to bind to and activate Src. Alternatively, sperm-egg fusion might introduce an already phosphorylated Src activator into the egg cytoplasm from the sperm. Either of these mechanisms would be consistent with the requirement for the Src SH2 domain in the interaction leading to the activation of the Src-like kinase in the egg at fertilization.