2016-10-28

The division Haptophyta is a lineage of unicellular algae that are widespread and often very abundant in diverse marine settings. Most haptophytes occur as solitary motile or non-motile forms, but a few form colonies or short filaments. Haptophyte cells are usually covered with one or several layers of organic scales of varying degrees of complexity, these being formed intracellularly in Golgi-derived vesicles. Haptophytes are characterized by the presence of a unique organelle called a haptonema (from the Greek Hapsis, touch, and Nema, thread), which is superficially similar to a flagellum but differs in the arrangement of microtubules and in function, being implicated in attachment or capture of prey. The haptonema is present in most species, sometimes in a reduced or vestigial form, but may rarely be absent.

The Haptophyta includes 2 classes: the Pavlovophyceae with only 13 described species and the Prymnesiophyceae which contains the vast majority of the known diversity of haptophytes and which comprises 2 orders of non-calcifying taxa, the Phaeocystales and the Prymnesiales, together with the calcifying coccolithophores making up a monophyletic clade (the sub-class Calcihaptophycidae) containing 4 orders (Isochrysidales, Coccolithales, Syracosphaerales, Zygodiscales).

There are some well-known non-calcifying haptophyte taxa, such as Phaeocystis, Prymnesium and Chrysochromulina, that form periodic harmful or nuisance blooms in coastal environments. However, the most familiar haptophytes are the coccolithophores, members of the Prymnesiophyceae that, in addition to a proximal layer of organic body scales, are covered with a distal layer of calcified scales (coccoliths) that are also formed intracellularly and that often have complex ornamentation. Coccolithophores are responsible for a large part of modern oceanic carbonate production and are thus key actors in global carbon cycling (Rost and Riebesell 2004).

Heteromorphic life histories have been documented in many members of the haptophyte class Prymnesiophyceae. These include alternations between non-motile and flagellated stages, between colonial and single cell stages, and between benthic and planktonic stages. The earliest studies on haptophyte life cycles focused mainly on members of the coccolithophores families Pleurochrysidaceae and Hymenomonadaceae (order Coccolithales), these being relatively easy to maintain in laboratory culture. Alternation of a non-calcifying (‘Apistonema’) stage with a coccolith-bearing stage has been reported in Ochrosphaera (Schwarz 1932; Lefort 1975), Hymenomonas(Fresnel 1994) and Pleurochrysis (Leadbeater 1970, 1971; Gayral and Fresnel 1983). In Pleurochrysis carterae (Rayns 1962) and Hymenomonas lacuna (Fresnel 1994), chromosome counting confirmed that the non-calcifying stage in these life cycles is haploid and the calcifying stage diploid, providing the first hard evidence of the existence of haplodiplontic life cycles in haptophytes (Fig. 6).



Figure 6

These life cycles were relatively easy to discern due to the presence of coccoliths (that are visible in light microscopy) in one of the phases. Two main types of coccoliths exist: heterococcoliths (formed of a radial array of complex-shaped interlocking crystals units) and holococcoliths (constructed of numerous small, similar sized and simple-shaped calcite elements). A culture study by Parke and Adams (1960) on the non-motile heterococcoliths bearing stage of Coccolithus braarudii (Coccolithales) demonstrated an alternation with Crystallolithus hyalinus, a motile stage bearing holococcoliths. Prior to this observation, heterococcolithophores and holococcolithophores had been considered as taxonomically discrete groups of species.

Reviewing these and other studies, Billard (1994) suggested that haptophyte life cycles typically include haploid and diploid phases, each capable of independent asexual reproduction (haplodiplonty), with distinct patterns of body scale ornamentation (and in some cases coccolith type) characteristic of each ploidy state. Prymnesiophycean body scales are composed of microfibrils and contain proteins and carbohydrates including cellulose (Leadbeater 1994 and references therein). The proximal (“body”) scales are composed of two layers with the proximal face (facing the cell membrane) having a radial pattern of microfibrils often arranged into quadrants, whereas the distal face either has a radial pattern or an interwoven spiral pattern of concentric rings. In this scheme, the body scales of the diploid cells have identical (radial) ornamentation on both sides, whereas those of the haploid stage have distinct patterns on the proximal and distal faces (radial and spiral, respectively). Billard (1994) predicted that the heterococcolith bearing phase of C. braarudii is diploid and the holococcolith-bearing phase haploid, by likening the patterns of body scale ornamentation in the known haplodiplontic life cycles of species in the Pleurochrysidaceae and Hymenomonadaceae with those in the life cycle of C. braarudii as illustrated by Manton and Leedale (1969). A number of other coccolithophores for which body scales have been illustrated in one phase only fi t this pattern, including the heterococcolithophores Syracosphaera pulchra (Inouye and Pienaar 1988), Umbilicosphaera hulburtiana (Gaarder 1970) and Jomonlithus littoralis (Inouye and Chihara 1983), and the holococcolithophores Calyptrosphaera sphaeroidea (Klaveness 1973) and Calyptrosphaera radiata (Sym and Kawachi 2000).

DNA quantification by flow cytometry later confirmed the haplodiplontic nature of the holococcolithophore-heterococcolithophore life cycle of Coccolithus braarudii as well as of two other species for which both phases were maintained in culture (Houdan et al. 2004). Further indirect evidence that Haplo-diplontic life cycles are widespread in coccolithophores comes from observations in field samples from various locations of ‘combination coccospheres’ bearing both heterococcoliths and holococcoliths, interpreted as capturing the instant of a life cycle phase change (Thomsen et al. 1991; Kleijne 1992; Alcober and Jordan 1997; Young et al. 1998; Cros et al. 2000; Cortes and Bollmann 2002; Geisen et al. 2002; Cros and Fortuno 2002). These observations indicate that life cycles with alternating heterococcolith-bearing (diploid) and holococcolith-bearing (haploid) stages span a large part of the diversity of coccolithophores.

Over time, culture studies and observation of combination coccospheres have demonstrated that diploid generations in coccolithophore life cycles always bear heterococcoliths, whereas the haploid generations are covered by either holococcoliths (Coccolithaceae, Calcidiscaceae, Helicosphaeraceae, Syracosphaeraceae), aragonitic coccoliths (Polycrater), nannoliths (Ceratolithaceae), a non-calcifying benthic stage (Pleurochrysidaceae, Hymenomonadaceae) or a non-calcifying motile stage (Noëlhaerhabdaceae) (Billard and Inouye 2004 and references therein).

Alternation between generations of different ploidy levels are known to occur in each of the other two non-calcifying orders within the Prymnesiophyceae. In the Phaeocystales, the most complete information on the life cycle has been obtained for Phaeocystis globosa (reviewed by Rousseau et al. 2007). A Haplo-diplontic life cycle has been described for this species, which includes diploid colonial cells (recorded either as free non-motile cells or within colonies), diploid flagellate cells without organic scales, and two types of haploid flagellates surrounded by organic scales, differing in size (meso- and microflagellates). P. Antarctica is thought to exhibit a similar life cycle (Zingone et al. 2011). In the Prymnesiales, Prymnesiumparvum has been shown to be the diploid stage in a life cycle in which the haploid stage was originally described as a separate species, P. patelliferum (Larsen and Medlin 1997; Larsen and Edvardsen 1998) and Prymnesium polylepi (=Chrysochromulina polylepis) has also been demonstrated to have a Haplo-diploid cycle (Edvardsen and Vaulot 1996; Edvardsen and Medlin 1998). In each of these cases, body scale ornamentation fits the scheme of Billard (1994) and in some cases morphological differences are also evident in the distal organic scales (e.g., Probert and Fresnel 2007). The details of these non-calcified scales can only be observed by electron microscopy, generally making life cycles in these non-calcifying haptophytes difficult to identify.

In summary, dimorphic Haplo-diplontic life cycles appear to be widespread in the Prymnesiophyceae. To date, alternation of generations has not been demonstrated in members of the other haptophyte class, the Pavlovophyceae. The species within this distinctive clade do not possess the ornamented plate scales which have often proved indicative of ploidy state in the Prymnesiophyceae, and given the fact that different ploidy stages in many non-calcified members of the latter class can only be morphologically distinguished by this character, it is perhaps not surprising that the potential existence of Haplo-diplontic life cycles in the Pavlovophyceae has not been recognized. Transitions from non-motile to motile cells are common in the Pavlovophyceae (Billard 1994; Bendif et al. 2011), but relative motility is not often a good indicator of ploidy state.

There are few reports of cysts in the Haptophyta. Cysts of Prymnesium were described by Carter (1937) and Conrad (1941) and have been investigated by Pienaar (1980) who has shown that the walls of Prymnesium parvum cysts are composed of layers of scales with siliceous material on the distal surfaces. It is not known whether formation of these cysts is related to ploidy level.

It should be noted that in haptophyte life cycles the existence and place of sexuality, if applicable, generally remains unknown (Billard 1994). In the coccolithophores, sexuality has been revealed by direct observation of syngamy in only three species, Ochrosphaera neapolitana (Schwarz 1932), Pleurochrysis pseudoroscoffensis (Gayral and Fresnel 1983) and Coccolithus braarudii (Houdan et al. 2004). In O. neapolitana, meiosis, isogamete formation and syngamy were reported by Schwarz (1932). From light microscope observations on Coccolithus braarudii (Houdan et al. 2004) and Pleurochrysis pseudoroscoffensis (Gayral and Fresnel 1983), certain general inferences can be made about the meiotic process in coccolithophore life cycles. In Coccolithus braarudii, meiosis may occur within the heterococcosphere prior to production of the flagellar apparatus and subsequent liberation of the motile cell. Since only one viable cell emerges this would imply the redundancy of the other haploid nuclei formed by the meiotic divisions. Meiosis in the chlorophyte Spirogyra (Harada and Yamagishi 1984) is one of the best-known examples of this pattern. Alternatively, the motile cell that emerges from the heterococcosphere may still be diploid, with nuclear reduction occurring in subsequent divisions. The observation that holococcolith production does not commence immediately after formation of the flagellar apparatus may be interpreted as providing support for this second hypothesis (Houdan et al. 2004). In the life cycle of P. pseudoroscoffensis, four non-calcified motile haploid cells are formed within a heterococcosphere and following release, these cells remain motile for a short time before settling and dividing asexually to initiate the haploid non-calcified pseudofilamentous stage (Gayral and Fresnel 1983). Comparable ‘meiospores’ are formed in the life cycles of other members of the Pleurochrysidaceae (von Stosch 1967; Leadbeater 1971) and the Hymenomonadaceae (Fresnel 1994). In coccolithophores, the mode of initial gamete attraction and contact is not known, but once initiated, syngamy can clearly be completed within a very short period of time. The rapid initiation of heterococcolith production in the zygote observed in C. braarudii by Houdan et al. (2004) and reported for P. pseudoroscoffensis by Gayral and Fresnel (1983) indicates that the two haploid nuclei must fuse immediately following cytoplasmic fusion. In both species, a complete heterococcosphere was formed within 24 hours of the onset of fusion. From this limited evidence, it appears that in coccolithophores fusing gametes are isogamous and are morphologically indistinguishable from vegetative haploid cells, and that fusion can occur within a clone (homothallism). To our knowledge, crossing experiments between haploid coccolithophore strains have never been attempted.

In Phaeocystis globosa, micro- and mesoflagellates (haploid stages) are produced (presumably by meiosis) within the colony and are eventually released and multiply vegetatively. The life cycle is completed by syngamy between a micro- and mesoflagellate that develops into a diploid macroflagellate that is believed to develop into a new colony. However, the formation of colonial stages was never observed in cultures of P. globosa containing only haploid cells (Vaulot et al. 1994). This could be explained either by assuming that conjugation only occurs between the two heteromorphic types of haploid flagellates (anisogamy) or that different mating types exist amongst haploid flagellates (heterothallism). A non-motile zygote linking the haploid unicellular stages and the diploid colonial stages has been documented in P. Antarctica (Gaebler-Schwarz et al. 2010). This zygote can divide vegetatively as a benthic palmelloid stage and not revert to the colonial stage at least in culture conditions.

Studies on the mechanisms of sexual reproduction in haptophytes are currently restricted by the limited number of cultures available and, moreover, by the lack of clear indications as to the factor(s) responsible for the induction of life cycle phase changes in this group. The transition between the life stages in haptophytes is presumably controlled by the interplay of endogenous and environmental factors, but the role and the relative importance of these factors are poorly known. A number of reports suggest that in the Pleurochrysidaceae factors such as temperature and light (Leadbeater 1970) and the addition of fresh medium (Inouye and Chihara 1979; Gayral and Fresnel 1983) may influence phase changes. In cultures of Calyptrosphaera sphaeroidea, Noel et al. (2004) demonstrated that exposure to selected vitamins and trace metals induced the transition to the heterococcolith-bearing phase, whereas a slightly higher concentration of components in the basic medium along with concomitant stresses of light and temperature induced formation of the holococcolith-bearing phase.

Houdan et al. (2004) suggested that concentrations of inorganic or organic trace elements in the medium may have played a role in phase change induction in three coccolithophore species, and also hypothesized that a biological clock may be involved in this process.

Life cycle transitions with each phase adapted to distinct ecological niches may also be an integral part of the ecological strategy of haptophytes. In ecological terms, a Haplo-diplontic life cycle is generally considered as an adaptation to an environment which is seasonally variable or that contains two different niches (see review by Valero et al. 1992). There is some evidence that heterococcolith-bearing and holococcolith-bearing or non-calcifying phases in the life cycle of certain coccolithophore species have differential in situ spatiotemporal distributions (Cros and Fortuno 2002; Frada et al. 2012). In general, holococcolith-bearing stages (which always possess flagella) may be adapted to warm stratified surface waters, whereas the more robust (and often non-motile) heterococcolith-bearing stages may be better suited to turbulent mixed-layer waters. Noel et al. (2004) extrapolated from growth medium preferences of the two stages of Calyptrosphaera sphaeroidea to propose a hypothetical ecological cycle in which the holococcolith-bearing stage occurs in offshore waters and the heterococcolith-bearing stage in coastal waters. Very few studies have experimentally compared the physiological characteristics of different ploidy stages in haptophytes, but in these studies differences in responses to physic-chemical and/or biotic parameters have consistently been found.

For example, in the extremely abundant coccolithophore Emiliania huxleyi, which has a life cycle with diploid non-motile heterococcolith-bearing cells alternating with haploid flagellate non-calcifying cells, the haploid stage appears to be relatively sensitive to high light (Houdan et al. 2005), but not susceptible to E. huxleyi specific viruses (EhVs) that routinely infect and kill diploid cells (Frada et al. 2008). Dramatic differentiation in gene expression between diploid and haploid phases of the coccolithophore E. huxleyi have been demonstrated, with greater transcriptome richness in diploid cells suggesting they may be more versatile for exploiting a diversity of rich environments whereas haploid cells appear to be intrinsically more streamlined (Von Dassow et al. 2009).

Show more