Sperm-induced calcium oscillations: Soluble factor, factors or receptors?

The two questions posed by Tesarik (1998) and Swann et al. (1998) on the nature of sperm factor are: (i) if glucosamine-6-phosphate deaminase is not the `so-called oscillogen', as shown by the recombinant studies of Wolosker et al. (1998), what is the nature of sperm factor?; and (ii) is there a dual mechanism for oocyte activation involving signals from soluble sperm factors together with signals generated by the interaction of the spermatozoon with receptors on the oocyte plasma membrane. The debate really started in Edwardian times when it was shown that cortical granule exocytosis and oocyte activation in sea urchins were triggered by a component of the spermatozoon that enters the oocyte cytoplasm (Robertson, 1912; Loeb, 1913). Contrasting models emphasized the interaction of the spermatozoon with the external surface of the oocyte membrane (Lillie, 1922; Mazia et al., 1975). In the early 1980s, direct experimentation re-kindled the controversy. Direct evidence for a soluble sperm factor was shown by microinjecting soluble components from spermatozoa into sea urchin and ascidian oocytes (Dale et al., 1985; Dale, 1988). Several activation events including cortical granule exocytosis and gating of plasma membrane currents were triggered. These initial studies were supported with experiments on mammals, where it was shown that activation events, including calcium oscillations, were triggered by soluble sperm extracts (Stice and Robl, 1990; Swann, 1990; Dale et al., 1996; Parrington et al., 1996). Further credibility was gained with the development of intracytoplasmic sperm injection (ICSI) in the 1990s, since injection of a spermatozoon directly into the oocyte, leading to normal embryogenesis and live birth, seemed to preclude the membrane receptor theory. Since sperm factor was shown to induce a series of calcium oscillations in mammalian oocytes, the term `oscillogen' was coined (Parrington et al., 1996). Before entering into the debate we would like to suggest that this term is inadequate and does not represent this physiological trigger that spans the animal kingdom. We re-propose the use of `soluble sperm factor' (Dale et al., 1985). There are several reasons for this. Firstly, the capacity for mammalian oocytes to generate oscillatory calcium transients does not depend on the spermatozoon but is an intrinsic oocyte characteristic acquired during oocyte maturation. In fact, sperm-induced activation in immature oocytes, or indeed in aged oocytes, triggers calcium release but in an irregular fashion (Homa, 1995). Secondly, many chemicals such as thimerosal and activators, or inhibitors, of protein kinase C will trigger calcium oscillations in mammalian oocytes (Swann, 1991; Souza et al., 1996). Thirdly, in oocytes from sea urchins and other animals, spermatozoa do not induce oscillatory calcium transients, but a large single calcium wave (Whitaker and Swann, 1993). Finally, in ascidians, although oscillatory calcium transients are induced by the spermatozoon or microinjection of soluble sperm factor (Wilding and Dale, 1998) these are not required for cell progression from the metaphase I block (Russo et al., 1996). Soluble sperm extracts can activate oocytes both from different phyla and from different species (Dale, 1988; Homa and Swann, 1994; Wilding et al., 1997; Wu et al., 1997). More recently, Wakayama et al. (1998) have demonstrated that it is possible to partially activate mammalian oocytes by microinjecting sea urchin spermatozoa into the cytoplasm. Thus sperm factors are neither species specific, nor indeed phylum-specific. Sperm extracts can also trigger calcium oscillations in somatic cells (Currie et al., 1992; Berrie et al., 1996), suggesting that they are common calcium-releasing agents and are not sperm-specific molecules. The mode of calcium release at fertilization varies from species to species. There are several categories of calcium release mechanisms in oocytes, depending on the type of receptor located on the intracellular calcium store. Firstly, there is inositol 1,4,5-trisphosphate (IP3)-induced calcium release (IICR), and secondly, calcium-induced calcium release (CICR) and finally, NAADP+-induced calcium release. IICR is triggered by the binding of IP3 to its receptor on the endoplasmic reticulum (Terasaki and Sardet, 1991). IP3 is produced by the action of phospholipase C on the plasma membrane lipid phosphatidylinositol bisphosphate (PIP2) (Berridge, 1993). CICR is triggered by the opening of the ryanodine receptor on an intracellular store, but can also be triggered in a mechanism involving the IP3 receptor (Endo, 1977; Berridge, 1996). This can be triggered by calcium itself, and appears to be modulated by cyclic ADP ribose (Galione and White, 1994). Cyclic ADP ribose is in turn produced by metabolism of nicotinamide adenine disphosphate (NAD+) by ADP ribosyl cyclase or NAD+ glycohydrolase (Galione and White, 1994, Jacobson et al., 1995; Lee et al., 1995). More recently, other calcium-releasing second messengers have been discovered including cATP ribose and NAADP+ (Lee and Aarhus, 1995; Genazzani and Galione, 1996; Zhang et al., 1996). Since NAD+, NADH, NADP+ and NADPH can be metabolized to calcium-releasing second messengers in sea urchin microsomes (Clapper et al., 1987) other calcium releasing second messengers may be discovered in the nicotinamide nucleotide family. In sea urchins oocytes, both IICR and CICR are triggered at fertilization (Whitaker and Swann, 1993). In frog oocytes, IICR appears to be uniquely activated at fertilization (Whitaker and Swann, 1993), however, what appears to be IICR may in fact be CICR through the IP3 receptor (Galione et al., 1993b, Whitaker and Swann, 1993). The urochordate Ciona intestinalis also releases calcium by both IICR and CICR at fertilization, and also generates repetitive calcium transients through meiosis I and II (Speksnijder et al., 1990; McDougall and Sardet, 1995; Russo et al., 1996). These transients appear to be triggered by an IP3-dependent mechanism since they are blocked by heparin (McDougall and Sardet, 1995; Russo et al., 1996). In mammalian oocytes at fertilization there is a large increase in the sensitivity to CICR together with a series of repetitive calcium spikes (Igusa and Miyazaki, 1983; Cuthbertson and Cobbold, 1985; Miyazaki, 1988; Kline and Kline, 1992; Taylor et al., 1993). This again suggests activation of both CICR and IICR at fertilization. Mammalian oocytes contain both ryanodine and IP3 receptors (Miyazaki et al., 1992; Rickfords and White, 1993; Ayabe et al., 1995; Yue et al., 1995; Sousa et al., 1996). However, it is not yet certain whether repetitive calcium transients in mammalian oocytes are propagated by IICR or CICR (Carrol and Swann, 1992; Miyazaki et al., 1992, 1993; Swann, 1992; Kline and Kline, 1994; Fissore et al., 1995; Tesarik et al., 1995; Berridge, 1996; Tesarik and Sousa, 1996). In conclusion, spermatozoa contain many calcium-releasing molecules, including cGMP, IP3, nicotinamide nucleotide metabolites, calcium ions and relatively large proteins (Iwasa et al., 1990; Whitaker and Crossley, 1990; Tosti et al., 1993; Parrington et al., 1996) and oocytes possess several calcium release mechanisms. Since the active fraction in spermatozoa is neither species-specific, nor gamete-specific, and since a common activation pathway for the various calcium release mechanisms has not yet been identified, it is feasible to assume that sperm factor is a collection of second messengers found in many cell types but packaged and delivered differently in spermatozoa. To date, the extraction techniques for soluble sperm factors have not proven to be very efficient. Attempts to isolate and purify cytosolic components from spermatozoa have been fraught with difficulty and artefacts. The sperm factors generated do not mimic the whole spectrum of activation events. For example in ascidians, sperm factor triggers calcium oscillations but does not lead to the extrusion of the first polar body, nor does it generate plasma membrane ion currents or make the plasma membrane refractory to other spermatozoa (Wilding and Dale, 1998). So what are the other factors involved? Tesarik (1998) has suggested that the first slow sperm-induced calcium wave in human oocytes is the result of the interaction of the spermatozoon with receptors on the surface of the oocyte, and that the subsequent calcium transients are the result of soluble sperm factor interacting with and sensitizing intracellular calcium stores. The c-kit tyrosine kinase receptor has been proposed as a candidate activating molecule found in spermatozoa, although its localization to the midpiece of the spermatozoon make this improbable (Sette et al., 1997). Microinjection of human or ascidian soluble sperm extract into the cytoplasm of ascidian oocytes triggers both the initial slow calcium wave followed by the faster repetitive calcium transients showing that the initial slow transient is triggered by sperm factor, not by an external receptor mechanism (Wilding et al., 1997; Wilding and Dale, 1998). In the ascidian oocyte, the slow initial calcium wave is propagated by CICR and IICR, while the subsequent transients are propagated only by IICR (Wilding and Dale, 1998). Several major events occur at fertilization in ascidian oocytes. These include a large inward membrane current, intracellular calcium oscillations and inactivation of maturation promoting factor (MPF) (Dale and DeFelice, 1990; McDougall and Sardet, 1995; Russo et al., 1996). Cyclin-dependent kinase 1 (cdk1) activity is maximal at metaphase I and metaphase II, and decreases at exit from meiosis I and meiosis II. A series of calcium oscillations occur simultaneously with the decrease in cdk1 activity at metaphase I exit, while a second group of intracellular calcium transients precedes the cdk1 increase at metaphase II (Russo et al., 1996). This calcium signalling system, however, does not appear to be the central cell cycle control mechanism during meiosis I. We have shown that inactivation of the main cell cycle control protein, cdk1, is independent of intracellular calcium at this stage (Russo et al., 1996). However, oocytes do not extrude a polar body after fertilization in the presence of calcium chelators, suggesting that calcium is involved in the completion of meiosis I. This suggests that there are a number of different signals involved during meiosis I progression. Among them, ADP ribose probably plays an important role in completion of meiosis I. Nicotinamide, an inhibitor of nicotinamide nucleotide metabolism (Sethi et al., 1996) blocks the large, initial calcium transient triggered at fertilization suggesting that nicotinamide nucleotides do not directly trigger intracellular calcium release, but enhance influx by opening the ADP ribose channel in the plasma membrane. In fact, we have shown that nitric oxide is released at fertilization in ascidian oocytes leading to the production of ADP ribose (Grumetto et al., 1997) which directly gates the fertilization channel (Wilding et al., 1998). These experiments show that, at fertilization in ascidian oocytes, multiple independent metabolic pathways are stimulated by more than one factor contained within the cytoplasm of the spermatozoon. The fertilizing spermatozoon fuses to the plasma membrane and releases sperm factors into the oocyte. These factors stimulate both the production of IP3 and ADP ribose, the latter probably through the production of nitric oxide. ADP ribose gates the fertilization channels, and may be involved in the inactivation of MPF at meiosis I release. IP3 gates the release of intracellular calcium, which is required for the activation of MPF and the completion of meiosis II. The nature of sperm factor is still a matter for conjecture. We put our bets on multiple soluble factors, possibly also a membrane bound component, that are common to other cell types but packaged uniquely in the male gamete to ensure a rapid delivery to the targets located in the oocyte cytoplasm.