Male infertility-linked point mutation disrupts the Ca²⁺ oscillation-inducing and PIP₂ hydrolysis activity of sperm PLCζ

During mammalian fertilization, the fertilizing spermatozoon initiates a series of biochemical and morphological events in the egg known as ‘egg activation’. In all species examined, the earliest egg activation event is an increase in the cytosolic free Ca²⁺ concentration [1,2]. In mammals, this Ca²⁺ signal is delivered in the form of long-lasting Ca²⁺ oscillations that commence after gamete fusion and persist beyond the completion of meiosis [3]. This Ca²⁺ signalling phenomenon is necessary and sufficient for the completion of all of the events of egg activation [4,5]. Much controversy existed over how the sperm induces this fundamental developmental event, but growing evidence supports the notion that, during mammalian fertilization, egg activation is triggered by a sperm-specific PLC (phospholipase C) isoform, PLCζ [6–9]. PLCζ introduced into the ooplasm is able to hydrolyse PIP₂ (phosphatidylinositol 4,5-bisphosphate) to yield IP₃ (inositol 1,4,5-trisphosphate), thus triggering Ca²⁺ oscillations within the egg via the IP₃ receptor-mediated Ca²⁺ signalling pathway [10]. PLCζ has the smallest molecular mass and most elementary domain organization among mammalian PLC isoforms [10,11]. PLCζ consists of a tandem pair of EF hand domains at the N-terminus, followed by catalytic X and Y domains, and a C-terminal C2 domain [6,10].

Further support for the importance of PLCζ in fertilization has arisen from two clinical reports demonstrating either a reduced protein level or mutated forms of PLCζ in cases of human male infertility [12,13]. One infertility case identified following failed IVF (in vitro fertilization) treatment was associated with a point mutation in the PLCζ catalytic Y domain [13], where replacement of histidine with a proline residue (H398P) correlated with the absence of Ca²⁺ oscillation-inducing activity of human PLCζ [13]. His³⁹⁸ is conserved in PLCζ from various mammalian species as well as in PLCδ1, the most closely related isoform to PLCζ [14]. In the present study, we have introduced the infertility-linked human PLCζ H398P mutation into the equivalent His⁴³⁵ residue of mouse PLCζ to give PLCζ^H435P (Figure 1A) and analysed the effect of this mutation on in vivo Ca²⁺ oscillation-inducing and in vitro PIP₂ hydrolysis activity. For comparative analysis, we also replaced His⁴³⁵ with a neutral non helix-destabilizing residue, alanine, to produce PLCζ^H435A. An additional charge-reversal mutant, PLCζ^D210R, which produces an inactive enzyme [6], served as a negative control. We also examined the effect on PIP₂ hydrolysis activity of replacing in PLCδ1 the equivalent conserved His⁵⁴² to yield PLCδ1^H542P. Furthermore, we investigated potential dominant-negative inhibitory effects of PLCζ^H435P on the Ca²⁺ oscillation-inducing activity of WT (wild-type) mouse PLCζ (PLCζ^WT) and mouse sperm.

Mouse PLCζ-luciferase in pCR3 [15] was subjected to site-directed mutagenesis (QuikChange II; Stratagene) to generate the PLCζ^H435P, PLCζ^H435A and PLCζ^D210R mutants. PLCζ^WT and mutants were amplified by PCR from the corresponding pCR3 plasmid using Phusion polymerase (Finnzymes) to incorporate a 5′ EcoRI site and a 3′ SalI site and were cloned into pGEX-6P1 (GE Healthcare). The primers used for amplification of WT and mutant PLCζ were: 5′-ACATGAATTCATGGAAAGCCAACTTCATGA-3′ (forward) and 5′-TAACGTCGACTCACTCTCTGAAGTACCAAAC-3′ (reverse). Similarly, rat PLCδ1 in pGEX-5X2 [15] was subjected to site-directed mutagenesis to generate PLCδ1^H542P. Following linearization of WT and mutant PLCζs, cRNA was synthesized using the mMessage Machine T7 kit (Ambion) and a poly(A) tailing kit (Ambion), as per the manufacturer's instructions.

Experiments were carried out with mouse eggs in Hepes-buffered saline [H-KSOM (Hepes-buffered potassium simplex optimized medium)] as described previously [15,16]. Female mice were superovulated by injection of hCG (human chorionic gonadotropin; Intervet). Eggs were collected 13.5–14.5 h later and maintained in 100 μl of H-KSOM under mineral oil at 37 °C. Egg microinjection was carried out 14.5–15.5 h after hCG administration [16]. IVF experiments were carried out as described previously [6].

All procedures were in accordance with the UK Home Office Animals Procedures Act and approved by the Cardiff University Animals Ethics Committee.

Mouse eggs were microinjected with cRNA encoding the particular PLC(s) mixed with an equal volume of 1 mM Oregon Green BAPTA [1,2-bis-(o-aminophenoxy)ethane-N,N,N′,N′-tetra-acetic acid]–dextran (Molecular Probes) in injection buffer (120 mM KCl and 20 mM Hepes, pH 7.4). All injections were 3–5% of the egg volume. Eggs were then maintained in H-KSOM containing 100 μM luciferin and imaged on a Nikon TE2000 microscope equipped with a cooled intensified CCD (charge-coupled-device) camera (Photek). Ca²⁺ was monitored for 4 h after injection by measuring Oregon Green BAPTA–dextran fluorescence with low-level excitation light from a halogen lamp. After Ca²⁺ measurements, eggs were monitored for luminescence by integrating light emission (in the absence of fluorescence excitation) for 20 min using the same intensified CCD camera. Fluorescence signals were 10–100 times that for luminescence. Ca²⁺ measurements were considered valid only if the egg was also luminescent. To estimate expressed protein levels from luminescence values, luminescent eggs collected in a tube containing PBS (137 mM NaCl, 2.7 mM KCl, 4.3 mM Na₂HPO₄·7H₂O and 1.4 mM KH₂PO₄, pH 7.4) with 1 mM MgATP+100 μM luciferin were lysed with 0.5% Triton X-100 and the steady-state light emitted (in c.p.s.) was calibrated with recombinant firefly luciferase (Sigma) to give the mean protein expression of each PLCζ-luciferase construct [15]. Lower levels of protein at the start of experiments were estimated by linear extrapolation of the luminescence calibration curve.

For GST (glutathione transferase)-fusion protein expression, Escherichia coli [Rosetta (DE3); Novagen], transformed with the appropriate pGEX plasmid, was cultured at 37 °C until a D₆₀₀ of 0.6, and then protein expression was induced for 18 h at 16 °C with 0.1 mM IPTG (isopropyl β-D-thiogalactopyranoside) (Promega). Cells were centrifuged at 6000 g for 10 min, resuspended in PBS containing 2 mM DTT (dithiothreitol) and protease inhibitor mixture (Roche) and were then sonicated four times for 15 s on ice. After centrifugation at 15000 g for 15 min at 4 °C, soluble GST-fusion proteins were purified by affinity chromatography using glutathione–Sepharose™ 4B following standard procedures (GE Healthcare). Eluted proteins were dialysed overnight (SnakeSkin 10000 molecular mass cut-off; Pierce) at 4 °C in 4 litres of PBS, and concentrated with centrifugal concentrators (10000 molecular mass cut-off; Sartorius).

PIP₂ hydrolytic activity of PLCζ^WT, PLCδ1^WT and mutant constructs was assayed as described previously [15]. The final volume of the assay mixture was 50 μl containing 100 mM NaCl, 0.4% sodium cholate, 2 mM CaCl₂, 4 mM EGTA, 20 μg of BSA, 5 mM 2-mercaptoethanol and 20 mM Tris/HCl buffer, pH 6.8. The PIP₂ concentration in the reaction mixture was 220 μM, containing 0.05 μCi of [³H]PIP₂. Assay conditions were optimized for linearity, requiring incubation for 10 min at 25 °C with 20 pmol of protein. Recombinant proteins were separated by SDS/PAGE and immunoblot analysis was performed as described previously [15]. Proteins were probed with a polyclonal anti-GST antibody (1:10000 dilution).

To assess the Ca²⁺ oscillation-inducing activity of the PLCζ^WT, PLCζ^H435P, PLCζ^H435A and PLCζ^D210R mutants and to verify that these constructs were faithfully expressed as functional proteins in cRNA-microinjected mouse eggs, we prepared PLCζ-luciferase fusion constructs to enable quantification of relative protein expression, as described previously [15]. PLCζ^WT-injected eggs displayed prominent Ca²⁺ oscillations, with the first Ca²⁺ spike occurring at a luminescence of 0.35 c.p.s., which we estimate is equivalent to a protein expression level of 35 fg/egg, similar to that reported previously [6]. In contrast, microinjecting cRNA encoding either PLCζ^H435P or PLCζ^D210R failed to cause any Ca²⁺ oscillations (Figure 1B), even after relatively high levels of protein expression (Table 1). However, microinjecting the PLCζ^H435A mutant caused Ca²⁺ oscillations in all injected eggs (Figure 1B), exhibiting a broadly similar potency to PLCζ^WT, with the first Ca²⁺ spike detected after protein expression of 0.73 c.p.s. (Table 1). These results show that the mouse PLCζ H435P mutation is inactive in mouse eggs, whereas the H435A mutation retains Ca²⁺ oscillation-inducing activity.

PLCζ^H435P and PLCζ^D210R were subcloned into pGEX-6P1 and purified as GST-fusion proteins. We also prepared the corresponding substitution H542P in PLCδ1 and expressed PLCδ1^WT and PLCδ1^H542P to enable comparative isoform analysis of in vitro PIP₂ hydrolysis activity. The optimal protein yield for PLCζ^WT, PLCδ1^WT and their mutants required maintaining cultures at 37 °C until a D₆₀₀ of 0.5, followed by induction of expression with 0.1 mM IPTG for 18 h at 16 °C. Figure 2 shows glutathione affinity-purified recombinant proteins analysed by SDS/PAGE (left-hand panels) and anti-GST immunoblot analysis (right-hand panels). The predicted molecular mass for GST–PLCζ and its corresponding mutants was 100 kDa (Figure 2A), whereas for GST–PLCδ1^WT and the H542P mutant it was 111 kDa (Figure 2B). The corresponding proteins with appropriate molecular masses were observed as the highest band in both the gel and immunoblot. The band at 26 kDa is consistent with cleaved GST, which, together with several other intermediate molecular-mass bands detected by the GST antibody, are the probable result of some degradation occurring through protein expression and purification.

The enzyme activity of GST-fusion proteins for PLCζ^WT, PLCδ1^WT and corresponding mutants were determined using the [³H]PIP₂ hydrolysis assay at 1 μM and 1 mM Ca²⁺ (Figure 2C), as described previously [15]. PLCζ^H435P and PLCζ^D210R were enzymatically inactive at both low and high Ca²⁺, in contrast with PLCζ^WT which showed a specific activity of 449±36 nmol/min per mg of protein at 1 μM Ca²⁺ and 382±47 nmol/min per mg of protein at 1 mM Ca²⁺. PLCδ1^WT displayed high activity at 1 mM Ca²⁺ with PIP₂ hydrolysis of 1385±42 nmol/min per mg of protein compared with 395±27 nmol/min per mg of protein at 1 μM Ca²⁺. In contrast, PLCδ1^H542P was completely inactive at 1 μM Ca²⁺ and retained only 16% of the PLCδ1^WT activity at 1 mM Ca²⁺ (222±36 nmol/min per mg of protein). These results show that the above His→Pro mutation completely inactivates the PIP₂ hydrolysis activity of both PLCζ and PLCδ1, whereas the His→Ala substitution in PLCζ retains in vivo function.

To investigate whether PLCζ^H435P can alter the Ca²⁺ oscillation-inducing activity of PLCζ^WT, we co-microinjected into mouse eggs an equal mixture of cRNA encoding luciferase-tagged PLCζ^H435P and PLCζ^WT (Figure 3, top panel). The co-injected cRNA produced Ca²⁺ oscillations with comparable properties with those observed with PLCζ^WT alone (Figure 1B and Table 1), showing a time to the first peak of ~30 min (Table 2). This suggests that the expression of PLCζ^H435P at similar levels to PLCζ^WT does not interfere with Ca²⁺ oscillations. To determine whether an excess of PLCζ^H435P was required to block PLCζ^WT-induced Ca²⁺ oscillations, we performed sequential cRNA microinjections; PLCζ^H435P was introduced first, followed after a period of 1 or 2 h by a second injection of PLCζ^WT (Figure 3, middle and lower panels). This protocol employing prior expression of PLCζ^H435P for 1–2 h followed by PLCζ^WT expression did not interfere with the induction of normal Ca²⁺ oscillations ~30 min after PLCζ^WT cRNA injection (Table 2). Table 2 shows that, in the double cRNA injection experiments, the cumulative luminescence for both PLCζ^H435P and PLCζ^WT at the time of first spike (7.06 and 18.68 c.p.s. for 1 and 2 h respectively) was much higher than for control PLCζ^WT alone (0.35 c.p.s.; Table 1). This suggests that an excess of PLCζ^H435Pprotein does not interfere with PLCζ^WT-induced Ca²⁺ oscillations. The estimated PLCζ^H435P protein level of 400 fg in the egg after 2.5 h (Figure 3 and Table 2), when the first spike is observed ~30 min after injection of PLCζ^WT, is well above the ~35 fg estimated to be required for Ca²⁺ oscillations with PLCζ^WT alone (Figure 1 and Table 1) [6,9]. This result therefore suggests that the PLCζ^H435P protein even when expressed at a 10-fold excess remains unable to block Ca²⁺ oscillation-inducing activity of PLCζ^WT. To examine whether normal sperm-induced Ca²⁺ oscillations are affected by the presence of PLCζ^H435P, IVF experiments with mouse sperm were performed. Even after high protein expression levels of PLCζ^H435P were achieved in the mouse eggs (>40 c.p.s.), there was no discernable effect on sperm-induced Ca²⁺ oscillations (Figure 4).

Accumulating evidence suggests that sperm-specific PLCζ is the probable agent that stimulates Ca²⁺ oscillations and consequent egg activation after sperm–egg membrane fusion. Identification of PLCζ in several different mammalian species indicates that PLCζ should play a pivotal role at fertilization in all mammalian kingdoms. The importance of PLCζ in mammalian fertilization has been supported by two clinical studies which linked PLCζ with some cases of human male infertility [12,13]. ICSI (intracytoplasmic sperm injection) is a powerful IVF technique that has been extensively employed to overcome many male infertility conditions such as severe oligospermia, asthenospermia and teratospermia [12]. A study [12] has identified a number of patients with repeatedly unsuccessful ICSI due to egg activation failure. Sperm from these patients was unable to initiate Ca²⁺ oscillations and this deficiency was associated with reduced or absent PLCζ expression in their sperm [12]. A second clinical study reported identification of a point mutation in the PLCζ gene of a patient with failed ICSI [13]. In this mutant PLCζ, His³⁹⁸ in the catalytic Y domain was replaced with a proline residue. Injection of infertile sperm or the mutant PLCζ from this patient failed to induce the typical pattern of Ca²⁺ oscillations in unfertilized mouse eggs [13]. The His³⁹⁸ residue is conserved not only in PLCζ of all mammalian species, but also in PLCδ1, the most closely related PLC isoform to PLCζ. Interestingly, His³⁹⁸ does not correlate with one of the five critical active-site residues within the catalytic domain of PLCδ1 (His³¹¹, Glu³⁴¹, Asp³⁴³, His³⁵⁶ and Glu³⁹⁰) that were identified previously by structural studies [17] and which are conserved in PLCζ.

In the present study, we introduced the infertility-linked human PLCζ H398P mutation into the equivalent His⁴³⁵ residue of mouse PLCζ (H435P; Figure 1A) and assessed the effects of PLCζ^H435P upon in vivo Ca²⁺ oscillation-inducing and in vitro PIP₂ hydrolysis activity. Our results indicate that the H435P mutation totally abolishes the ability of PLCζ to trigger Ca²⁺ oscillations in mouse eggs (Figure 1B and Table 1) and also fully abrogates enzyme activity in a PIP₂ hydrolysis assay. To examine whether the His⁴³⁵ mutation to a proline residue, which in protein structure prediction is known to have helix-destabilizing ability, specifically causes sufficient perturbation of the catalytic Y domain to annul PLCζ enzymatic activity, we employed an additional PLCζ mutant in which His⁴³⁵ was replaced with a neutral non-helix-destabilizing amino acid, alanine. Interestingly, in contrast with the inactive PLCζ^H435P, robust Ca²⁺ oscillations were generated by PLCζ^H435A cRNA injection with a similar potency to PLCζ^WT (Figure 1B and Table 1). This suggests that the H435P substitution may cause inactivation of PLCζ due to the introduction of major protein structural changes, consistent with that proposed previously [13]. We examined further the effect of the PLCζ^H435P equivalent mutation upon the in vitro enzymatic properties of a closely related PLC isoform PLCδ1. We observed very similar results with the equivalent PLCδ1^H542P mutant, which retained only 16% of the PIP₂ hydrolysis activity of the PLCδ1^WT at high Ca²⁺ levels, a condition where PLCδ1^WT is known to be fully enzymatically active [11,15].

Interestingly, Heytens et al. [13] reported that the H398P mutation identified to correlate with male infertility appeared to be heterozygous, suggesting that in humans PLCζ^H398P may exert a dominant-negative effect on PLCζ^WT. To investigate whether mouse PLCζ^H435P acts in a dominant-negative fashion, we performed both cRNA co-microinjection and sequential injection experiments using PLCζ^H435P and PLCζ^WT. The aim of these experiments was to determine whether an equivalent or excess level of PLCζ^H435P protein expressed in mouse eggs can block the Ca²⁺ oscillation-inducing activity of PLCζ^WT. However, when PLCζ^H435P was expressed at up to 10-fold higher levels relative to PLCζ^WT, this did not block the ability of PLCζ^WT to trigger Ca²⁺ oscillations, suggesting that, in the mouse, PLCζ^H435P does not display dominant-negative behaviour. This observation is intriguing as it leaves unexplained why the sperm of this heterozygous infertile male is unable to cause Ca²⁺ oscillations after ICSI. Further analysis is required to confirm whether there are any other defects, such as PLCζ mislocalization, in the sperm of this infertile patient. Transgenic heterozygous animal models carrying this mutant allele would also help to enable further investigation of the reason for failure of egg activation in this clinical case.

The present study extends the previous work of Heytens et al. [13] by revealing that (i) a mouse PLCζ H435P mutation, equivalent to the infertile human H398P, is functionally inactive both in vivo and in vitro, as is the equivalent PLCδ1 mutation H542P, and (ii) injection of luciferase-tagged mutant PLCs enables quantitative analysis of their in vivo expression in eggs and the demonstration that a dominant-negative inhibition mechanism by the mouse H435P mutant PLCζ does not appear to operate. Further advances in understanding the precise role and importance of PLCζ in mammalian fertilization may provide a major step in overcoming some cases of male infertility. It has been reported that, in some cases of egg activation failure after ICSI, treatment with a Ca²⁺ ionophore has been successful in producing assisted egg activation [18]. However, it remains to be determined whether routine ionophore treatment to assist with activation carries a risk of abnormal embryo development, or whether it is the most effective means of overcoming activation failure [19]. The application of recombinant PLCζ might represent a potential alternative physiological therapeutic agent that can overcome certain cases of failed fertilization after ICSI.

BAPTA

1,2-bis-(o-aminophenoxy)ethane-N,N,N′,N′-tetra-acetic acid

CCD

camera, charge-coupled-device camera

GST

glutathione transferase

hCG

human chorionic gonadotropin

H-KSOM

Hepes-buffered potassium simplex optimized medium

ICSI

intracytoplasmic sperm injection

IP₃

inositol 1,4,5-trisphosphate

IPTG

isopropyl β-D-thiogalactopyranoside

IVF

in vitro fertilization

PIP₂

phosphatidylinositol 4,5-bisphosphate

PLC

phospholipase C

WT

wild-type

Michail Nomikos, George Nounesis, Karl Swann and Anthony Lai devised the project strategy; Michail Nomikos and Anthony Lai designed the experiments, which were performed by Michail Nomikos, Khalil Elgmati, Maria Theodoridou, Brian Calver and Bevan Cumbes; and Michail Nomikos, Karl Swann and Anthony Lai prepared the manuscript.

We are grateful to Matilda Katan (Institute of Cancer Research, London, U.K.) for providing the PLCδ1, and to John Parrington (Department of Pharmacology, Oxford University, Oxford, U.K.) for helpful discussions.

This work was supported by the Wellcome Trust [grant number 080701] and the Medical Research Council [Developmental Pathway Funding Scheme (DPFS)]. K.E. and M.T. are research scholars supported by the Libyan Government and NCSR Demokritos respectively.