Reproduction in Dinoflagellates

Dinoflagellate sexual reproduction has long been disputed, and a 1973 textbook on protozoology contained no reference to this phenomenon (Grell 1973). The earliest documented report of sexual reproduction in dinoflagellates was Joseph’s (1879) description of pairing and fusion of swimming cells of Peridinium stygium, but careful studies by von Stosch in the 1960s (1965; 1969; 1972; 1973) have since transformed our understanding of dinoflagellate sexuality. With researchers’ increasing capacity to maintain laboratory cultures, sexuality has now been documented for some 100 species (Walker 1984; Wall and Dale 1968; Blackburn et al. 1989; Blackburn and Parker 2005). The reasons that sexual reproduction in dinoflagellates had so long been overlooked include: (1) gametes can look similar to vegetative cells; (2) gamete fusion is easily confused with cell division; (3) ‘warty’ zygotes have often been interpreted as aberrant cells (Pfister and Anderson 1987). To date sexual life cycles are increasingly elucidated using nuclear staining and flow cytometric techniques, not only in culture but also field surveys.

 

Life Cycle

The life cycle of Gymnodinium catenatum is representative of those observed in many dinoflagellates (Fig. 4). Motile vegetative cells divide vegetatively (by mitosis) to form chains. With the onset of sexual reproduction, vegetative division results in single cell gametes. Usually two types of gametes from different clonal strains (heterothallism) are required for sexual reproduction, pairs of which fuse to give a planozygote. This cell loses motility to form a benthic resting cyst (hypnozygote). Excystment produces a planomeiocyte, similar to a planozygotes, which divides (by meiosis) to re-establish the planktonic vegetative stage (after Blackburn et al. 1989). In Spanish cultures studied by Figueroa et al. (2006a, 2008) (dotted lines in Fig. 4), most planozygotes divided by binary fission to produce vegetative cells but without undergoing a cyst stage. Furthermore, some fusing gamete pairs did not form planozygotes but went through a division process before completing cytoplasmic fusion.

Figure 4

Patterns of Sexual Reproduction

All dinoflagellates studied to date, but one, exhibit the haplontic type of life cycle, the dominant vegetative stage (the one undergoing vegetative growth) being haploid. The single exception is Noctiluca which is claimed to have a diplontic life cycle (Zingmark 1970). Asexual reproduction can happen much more quickly, and therefore is the predominant manner of reproduction during optimal environmental conditions, but sexual reproduction is essential for species adaptation and allows for genetic recombination. Under appropriate conditions, dinoflagellate gametes are produced and fuse to form a diploid planozygote. Often gametes swim faster, can be paler in color and collect in ‘dancing groups’. Fusing pairs of gametes can be distinguished from dividing vegetative cells because their cingula are perpendicular. Dinoflagellates often produce gametes that do not differ morphologically from vegetative cells (a condition called hologamous). Fusing gametes can be identical to each other (isogamous) or be different from each other (anisogamous, e.g., Ceratium cornutum, C. horridum, Alexandrium tamarense). As for diatoms, species are monoecious or homothallic, i.e., sexual reproduction can occur within a clone (e.g., Alexandrium taylori), or dioecious or heterothallic, i.e., two different mating types (designated plus or minus) must be combined. Sexual compatibility can comprise only two different mating types (simple heterothallism), such as in Lingulodinium polyedrum (Figueroa and Bravo 2005b) or more complex heterothallism, such as in Alexandrium minutum (Figueroa et al. 2007). Sexual mating compatibility has been used to elucidate species synonymy but also genetic affinities between geographic populations of the same species (Blackburn et al. 2001).

 

Three types of zygotes have been reported: (1) planozygote motile throughout and meiosis is completed without cyst formation (e.g., Ceratium horridum); (2) planozygote loses motility and forms a temporary cyst (e.g., Helgolandinium subglobosum); (3) planozygote forms a resting cyst or hypnozygote. The planozygote can be identified by two (‘ski track’) longitudinal flagella. This stage often develops into a resting cyst with a thick resistant cell wall and often requires a period of dormancy before germination is possible. Studies on freshwater dinoflagellates in particular have indicated that the timing of meiotic division in the sexual phase is variable (Pfister 1975, 1976, 1977). Meiosis occurs (1) within the planozygote; or (2) after excystment to release an appropriate number (2–4) of daughter cells; or (3) by subsequent divisions of the single meiocyte released by excystment. In a growing number of dinoflagellate zygotes, the nucleus has been observed to enlarge further and rotate rapidly within the cell (so called ‘nuclear cyclosis’ first described by Pouchet 1883) which von Stosch (1973) associated with the onset of meiosis.

 

Until recently the most common pathway observed was the transition of planozygote to resting cyst but it is now thought that the planozygote can also skip cyst formation (Figueroa and Bravo 2005a,b). Other possible pathways are: (1) gametes can revert to an asexual phase and undergo binary fission rather than fusion (e.g., Gymnodinium nolleri, G. catenatum,

Alexandrium taylori or Lingulodinium polyedrium); and (2) planozygotes undergo meiosis and division without the production of a hypnozygote (Figueroa and Bravo 2005a, b; Figueroa et al. 2006a, b). In some species for which a sexual cycle has been reported no resting cyst is known (e.g., Karlodinium venefi cum, Karenia brevis; Walker 1982). Asexual resting cysts are also known, e.g., in Scrippsiella hangoei (Kremp and Parrow 2006). Another type of quiescent stage is what is variably called temporary, pellicle or ecdysal cysts with a thin wall and limited capacity to withstand adverse environmental conditions, either produced sexually or asexually.

 

Cyst as Survival Strategies

Interest in dinoflagellate sexual reproduction was triggered in the early 1960s with the recognition that many fossil cysts (first described by Ehrenberg 1838 from Cretaceous flints as ‘hystrichosphaeres’) are in fact dinoflagellate hypnozygotes. Wall and Dale (1968) conducted the first experiments incubating living ‘cysts’ from Woods Hole bottom sediments.

Excellent cyst preservation in the fossil record is due to the presence of sporopollenin in the wall of many (but not all) species (Fig. 5). Currently more than 80 marine and 15 freshwater species of modern dinoflagellates are known to produce resting cysts. This number of cyst-producing species is small however compared with the total number of extant dinoflagellates (more than 2000). Resting cysts can survive harsh environmental conditions and thus play an important ecological role as the inoculum for recurrent blooms. Dinoflagellate cysts can remain viable in sediments for up to 100 yrs. They also facilitate expansion of the geographical distribution of a species through cyst dispersal via ocean currents and even ship ballast water discharge.

 

Factors Triggering Sexual Reproduction

Sexuality has been traditionally achieved in culture through nutrient depletion, with temperature and light being important modulators of the cyst yield. However, gamete pairing and planozygote formation in nature may not be always linked to nutrient shortage, since sexuality has been observed either at termination of blooms or during active growth, or when reaching a cell abundance threshold (Garcés et al. 2002).

The conditions that trigger cyst formation and cyst germination have become the key focus of predicting blooms of cyst forming toxic dinoflagellates such as Alexandrium tamarense, Gymnodinium catenatum and Pyrodinium bahamense (Anderson and Wall 1978; Dale 1983). The dormancy period is a maturation time during which biological activity is suspended; this can last from hours to days (Kryptoperidinium foliaceum), weeks (Gymnodinium catenatum) to months (Alexandrium tamarense). Germination cannot be induced during dormancy, but once completed the cyst enters quiescence, during which germination can occur if environmental conditions are suitable. A genetic (endogenous) control of dormancy has been documented in some species (Anderson and Kiefer 1987), but exogenous (abiotic and physiological) modulators and internal clocks (endogenous rhythms) also play a role. Germination patterns thus can drive bloom strategies and seasonal species succession. Once germination occurs, the size of the inoculum will be influenced not only by the number of germinating cysts but also by their viability.

Figure 5 Sexually Produced Dinoflagellate Resting Cysts from Top Left to Bottom Right