Recombinant Phospholipase C-zeta Has High Ca2+-sensitivity and Induces Ca2+ Oscillations in Mouse Eggs*

used PLC and


INTRODUCTION Fertilized mammalian eggs exhibit repetitive transient rises in [Ca 2+ ] i called Ca 2+
oscillations due to repeated Ca 2+ release from the ER mainly through the InsP 3 receptor (1). The [Ca 2+ ] i rises are a pivotal signal for egg activation and responsible for early embryogenesis (2,3). Accumulated evidence suggests that Ca 2+ oscillations are induced by cytosolic sperm factor introduced into the ooplasm upon sperm-egg fusion (2,4).
Therefore, the identification of the Ca 2+ oscillation-inducing sperm factor, that is the eggactivating sperm factor, is a current central subject to elucidate the mechanisms of fertilization. Recently, Saunders et al. (5) have reported a novel type of PLC (the enzyme that produces InsP 3 and diacylglycerol from membrane PtdInsP 2 ), PLCζ that is specifically expressed in the mammalian sperm. Injection of RNA encoding PLCζ into mouse eggs causes Ca 2+ oscillations and subsequent early embryonic development, by expressed PLCζ at an estimated level comparable to the content in a single sperm (5). The Ca 2+ oscillationinducing activity of sperm extract (4,6) is lost when pretreated with an antibody against PLCζ (5). Thus, PLCζ is considered as a strong candidate of the sperm factor. To assure this possibility, it is primarily necessary to examine whether purified PLCζ protein induces Ca 2+ oscillations in the egg.
It has been shown that PLCβ1 (7), γ1 (8,9), γ2 (8), δ1 (10) and δ4 (7) are expressed in the mammalian sperm and that recombinant PLCβ1, γ1, γ2 and δ1 failed to cause Ca 2+ release in the ooplasm (11). PLCζ is the smallest PLC identified to date, lacking the Nterminal PH domain ( Fig. 1A) (5) that is found in all isoforms of PLC β, γ and δ, and is the site for interaction with membrane phospholipids (12). Since PLCζ as well as PLCδ lacks a regulatory domain such as the G protein-binding site of PLCβ or the SH domain of PLCγ for phosphorylation by tyrosine kinase, the activation mechanism of PLCζ and PLCδ is unknown. Therefore, it is also necessary to access how PLCζ undergoes the active state for production of InsP 3 . Here we first show that recombinant PLCζ protein induces Ca 2+ oscillations in mouse eggs and that PLCζ possesses an extremely high Ca 2+ sensitivity in the PtdInsP 2 -hydrolyzing activity to be active even at the resting state of

RESULTS AND DISCUSSION
PLCζ and PLCδ1 with N-terminal His tag were purified up to a major single band shown by silver staining (Fig. 1B). The molecular weight including His tag was ~78 and ~90 kDa, consistent with the full length PLCζ (647 amino acids) (5) and PLCδ1 (756 amino acids) (16), respectively.
Exogenously applied PLC is predicted to produce InsP 3 continuously in the egg and, thereby, cause repetitive Ca 2+ release from the ER. Microinjection of recombinant PLCζ into mouse eggs did induce Ca 2+ oscillations ( Fig. 2A) (n=10 eggs). The first Ca 2+ transient lasting for 4-5 min was followed by sharp Ca 2+ spikes at intervals of 2-3 min.
The interspike interval was progressively prolonged. The pattern is very similar to that of produced only a single Ca 2+ transient after a time lag of ~2 min ( Fig. 2B; n=3). This Ca 2+ response pattern is usually observed upon injection of diluted sperm extract (data not shown). The critical PLCζ concentration for induction of a single Ca 2+ release was between 15 and 30 µg/ml in the injection solution (Fig. 2, B and C).
A protein similar to but shorter than PLCζ (s-PLCζ) is expressed in the mouse testis (AK006672 in EMBL). s-PLCζ lacks 110 amino acid residues from the N-terminus corresponding to EF1, EF2, and EF3 domains, but is identical to PLCζ in EF4 and the succeeding region (Fig. 1A). s-PLCζ gene has not been found in the mouse genome and s-PLCζ is probably a splicing variant of PLCζ. Recombinant s-PLCζ elicited no Ca 2+ spike at the concentration 37-fold higher than that of PLCζ (Fig. 2D), indicating that EF domain(s) is significant in the induction of Ca 2+ oscillations.
Recombinant PLCδ1, which possesses basically similar domain features to those of PLCζ except the PH domain (Fig. 1A), induced Ca 2+ oscillations at the concentration of 1.2 mg/ml in the injection solution (30 µg/ml in the egg; Fig. 3A; n=6) and only a few Ca 2+ transients at 0.8 mg/ml ( Fig. 3B; n=3). Thus, about 20-fold higher concentration was required for induction of Ca 2+ oscillations, compared with PLCζ. PLCδ1 induced no Ca 2+ response at 0.6 mg/ml (Fig. 3C) or at the lower concentration range at which PLCζ induced Ca 2+ oscillations (data not shown).
The PLC activity (PtdInsP 2 -hydrolysing activity) of PLCζ in vitro was measured at [Ca 2+ ] between 10 -9 and 10 -4 M, in comparison with that of PLCδ1 (Fig. 4). No PLC activity was detected for PLCζ∆EFX at any [Ca 2+ ] (Fig. 4B), indicating that the samples from Sf9 cells had no intrinsic PLC activity. The assay was performed using the amount of PLC that was capable of generating maximal [ 3 H]InsP 3 of approximately 4,000 dpm, which corresponded to 0.32 n mol InsP 3 , under the condition that retained the linearity of InsP 3 formation to the reaction time and the enzyme concentration during incubation for 5 min. The amount used was 50 ng for PLCζ and 1 ng for PLCδ1 in 50 µl reaction mixture, corresponding to 1 µg/ml and 0.02 µg/ml, respectively. This concentration of PLCζ was in the same range as the intracellular concentration of injected PLCζ that induced Ca 2+ release (Fig. 2, A and B). The specific activity of PLCζ was 1.3 µ mol/mg/min at 1 µM Ca 2+ and that of PLCδ1 was 65 µ mol/mg/min at 30 µM Ca 2+ (Fig. 4,   A and B). The specific activity was 50-fold higher for PLCδ1. Nevertheless, PLCζ induced Ca 2+ oscillations at 20-fold lower concentrations than PLCδ1 ( Figs. 2A and   3A). This implies that PLCζ has 1000-fold higher efficiency than PLCδ1 in induction of Ca 2+ oscillations in the egg, when calculated on the basis of the specific PLC activity in vitro under the assumption of comparable dependence on [Ca 2+ ].
To address advantageous characteristics of PLCζ for induction of Ca 2+ oscillations, the dependence of PLC activity on [Ca 2+ ] was examined in vitro. The PLC activity of recombinant PLCδ1 was substantially recognized at 1 µM Ca 2+ (Fig. 4A). The activity steeply increased between 1 and 30 µM Ca 2+ and attained a saturation level at 30 µM Ca 2+ . The [Ca 2+ ] for giving a half maximal PLC activity, EC 50 , was obtained by fitting a curve to the data using Hill equation. It was calculated as 5.7 µM (Hill constant, 1.7). In contrast, the PLC activity of recombinant PLCζ was significantly recognized at [Ca 2+ ] as low as 10 nM, and reached a maximum at 1 µM Ca 2+ (Fig. 4B). EC 50 was 52 nM (Hill constant, 0.9); that is, ~100-fold lower than that for PLCδ1. It should be noted that PLCζ had 70% activity of the maximal level at 100 nM Ca 2+ which is the resting [Ca 2+ ] i level in mouse eggs (6,18) as well as somatic cells (19). This implies that PLCζ can be active even at the resting state of cells. PLCζ has the highest Ca 2+ -sensitivity among PLC isoforms identified and characterized to date (20)(21)(22)(23)(24).
[Ca 2+ ] i increases from 100 nM up to 500-1000 nM in each Ca 2+ spike during Ca 2+ oscillations in fertilized mouse eggs (25). In Figure 4B to give the ability to trigger Ca 2+ release in cells at the resting state. The two properties of PLCζ described above are appropriate for a Ca 2+ oscillation-inducing sperm factor, since the sperm factor is thought to be introduced into the ooplasm upon sperm-egg fusion, first triggers Ca 2+ release, and maintain Ca 2+ oscillations (2,4). It has been shown that boar sperm extract possesses one-third of the maximal PtdInsP 2 hydrolyzing activity at 100 nM Ca 2+ (27). Thus, sperm-specific PLCζ is a strong candidate of the mammalian eggactivating sperm factor. There might be an inactivation mechanism for PLCζ activity in the sperm before it is introduced into the ooplasm.
PLCδ1 possessed the much higher enzymatic activity in vitro, but had the much lower Ca 2+ oscillation-inducing activity in vivo, compared with PLCζ. PLCδ1 has been shown to be expressed in mouse immature germ cells, spermatogonia, but not detected in differentiated spermatides and spermatozoa (10). Thus, PLCδ1 is unlikely to be the sperm factor. The superiority of PLCζ to PLCδ1 in the Ca 2+ oscillation-inducing activity is thought to be derived from the much higher Ca 2+ -sensitivity (~100-fold difference in EC 50 ). However, the large difference in the efficiency of inducing Ca 2 + oscillations is not interpreted only in terms of Ca 2+ -dependent PLC activity in vitro, and suggests additional advantage of PLCζ. There might be an egg factor that promotes the activation of PLCζ, or special target membranes for PLCζ in the ooplasm. Further studies are necessary to determine whether PLCζ is the physiological sperm factor at fertilization as well as to elucidate the activation and modification mechanism of PLCζ on the basis of the molecular structure.    14 by guest on March 21, 2020