Comparative development of the spermatozoids of cycads and Ginkgo biloba.

  I. Abstract
 II. Introduction
III. Material and Methods
 IV. Observations and Discussion
     A. Spermatogenous Cell
     B. Blepharoplast
     C. Spermatid
     D. Morphology of the Spermatozoid
  V. Conclusions
 VI. Literature Cited

II. Introduction

Cycads and Ginkgo biloba L. are the only extant seed plants that produce flagellated male gametes. At least superficially these cells are quite similar in structure and action, and in both forms the motile structures (the flagella) are initiated from a multicentriolar body known as the "blepharoplast." Blepharoplasts are found in all archegoniate plants having motile sperm and arise apparently de novo, as described, for example, in Zamia (Webber, 1901; Turner, 1966), in Microcycas (Caldwell, 1907), in Marsilea (Mizukami & Gall, 1966), and in Ginkgo (Gifford & Larson, 1980); and they are remarkably similar both in ultrastructure and in function. We retain the original definition of the organelle as "a multicentriolar body" as described by Chamberlain (1898) but follow Gifford and Lin (1974, 1975) in referring to the centriole-like units forming the cortex of the blepharoplast as "probasal bodies." Blepharoplasts and developmentally related subcellular organelles such as basal bodies, flagella, microtubules, and a complex fibrous body known as the "multilayered structure" (MLS) have been observed by electron microscopy and described by several authors (e.g., in Ginkgo by Gifford & Lin, 1975; Gifford & Larson, 1980; Li et al., 1989; in Zamia by Norstog, 1974, 1975, 1986; and in Microcycas by Norstog 1990, 1993, 1995). These studies have described the origin of blepharoplasts (in Ginkgo only), the relationships in cycads and Ginkgo of blepharoplast probasal bodies to centrioles and flagellar basal bodies (Figs. 1-4 and 8-10), the formation of the MLS and the deployment of basal bodies and flagella in the maturing sperm (Figs. 4-7). They form the body of observations on which the following discussion is based.

[FIGURES 1-4 OMITTED]

III. Material and Methods

For Ginkgo, ovules in which male gametophytes were growing, were collected in July, August, and early September from Ginkgo trees on the Berkeley campus of the University of California and from trees in Vacaville, California. Microgametophytes were dissected from them and fixed in 4% glutaraldehyde in 0.15M phosphate buffer (pH 6.8) at room temperature for three hours or overnight, followed by several rinses in buffer, then by postfixation for one hour in buffered 1% Os[O.sub.4]. After rinsing in buffer, specimens were dehydrated in ethanol, embedded in an epoxy resin, sectioned, mounted on grids, stained with uranyl acetate and lead citrate, and examined and photographed with an electron microscope.

For cycads, ovules were collected from plants in the living collection at the Fairchild Tropical Garden in Miami, Florida. Collections for Zamia integrifolia were made in late May and early June and for Microcycas calocoma in February. Several collections were variously fixed in schedules of glutaraldehyde and Os[O.sub.4] as described in Norstog, 1974, 1986, and 1990. The resulting thin sections on grids were stained in uranyl acetate and lead citrate, and then examined and photographed using electron microscopes in the Departments of Biology at Northern Illinois University, the University of Miami, and Florida International University, Miami.

IV. Observations and Discussion

Spermatogenesis in Ginkgo biloba does not appear to be fundamentally different from that of Zamia integrifolia, these taxa being the subjects most intensively investigated both by light microscopy (Ikeno & Hirase, 1897; Webber, 1901; Lee, 1955; Favre-Duchartre, 1956) and electron microscopy (Norstog, 1974, 1975, 1986; Gifford & Lin, 1975; Gifford & Larson, 1980; Friedman, 1987; Friedman & Gifford, 1988, 1997; Larson & Gifford, 1988). However, in certain details there are marked differences, as, for example, in the early development of spermatogenous cells of Ginkgo and Zamia.

A. SPERMATOGENOUS CELL

The spermatogenous cell of Ginkgo differs markedly from those of cycads. In it a lens-shaped nucleus is flanked by a pair of globular bodies of unknown function, which Gifford and Lin (1975) simply call osmiophilic bodies. No such bodies have been observed in spermatogenous cells of cycads, and their nuclei are not lens shaped but spherical (see Webber, 1901, fig. 20; Norstog & Nicholls, 1997, figs. 3.44-3.45).

B. BLEPHAROPLAST

Spermatogenous cells of Ginkgo and cycads are similar in possessing a pair of multicentriolar bodies, one on each side of the nucleus. These generally are referred to as "blepharoplasts," a term coined by Webber (1901) specifically for the multicentriolar body of the Zamia spermatogenous cell that gives rise to the flagellature. It has been used in the same sense in Marsilea by Mizukami and Gall (1966), Bell (1974), and Hepler (1976), in Equisetum by Duckett (1973), in Ginkgo by Gifford and Lin (1975), and in Zamia by Norstog (1986).

During spermatogenesis in Ginkgo the blepharoplast, which in this taxon is a spherical body 3.5-4.5 [micro]m in diameter, becomes irregularly contorted (Figs. 1, 2) but retains its electron-dense interior appearance during and after mitosis until its probasal bodies become dissociated during mitosis of the spermatogenous cell (Fig. 3). These then become associated with a dense body that may function as a microtubule-organizing center (MTOC) (Fig. 4). During this period the developing MTOC is associated with the membranes of the nucellar beak, an outgrowth of the nucellar membrane of each spermatid (Figs. 5, 6). It subsequently participates in the formation of a fibrous body composed of fins, plates, microtubules, and flagellar basal bodies variously named in the literature but presently by consensus called "MLS."

[FIGURES 5-6 OMITTED]

The blepharoplast of the Zamia spermatogenous cell is comparatively large, about 26 [micro]m in diameter. That of Microcycas is much smaller, about 7 [micro]m in diameter, but the two are very much alike in appearance, as shown in Fig. 7. Fully formed blepharoplasts of Zamia and Microcycas resemble those of Ginkgo (about 3.5-4.5 [micro]m), except for being considerably larger and with more probasal bodies. Except in size, they all are alike in their electron-dense interiors interspersed with electron-lucent, vacuole-like areas in which occasional fibrous inclusions may be observed. Perhaps the latter are MTOCs that persist after the dissolution of the blepharoplasts and act as organizing centers for the MLS in Ginkgo, as in Figs. 3 and 4, and in Zamia, as in Fig. 10.

[FIGURES 7-10 OMITTED]

C. SPERMATID

Perhaps the most striking structure in the spermatid of Ginkgo is the aforementioned nucellar beak, an elaboration of the nucellar envelope that connects the nucleus with the cell's plasma membrane (see Favre-Duchartre, 1956, fig 55b, p. 118) and, shortly after losing this connection, initiates development of the MLS (Favre-Duchartre, 1956, fig. 55c, p. 118, and our Fig. 5). The very comprehensive and accurate light microscopy study of spermatogenesis in Zamia (Webber, 1901) failed to find any similar elaboration of the nucellar envelope or a nucellar beak, despite looking specifically for this structure. Perhaps more significantly, electron-microscopic studies of Zamia spermatogenesis have not revealed a nucellar beak structure (Norstog, 1986). The nucellar beak in Ginkgo apparently has a function relating to the organization and orientation of an MTOC, which in turn initiates formation of a MLS that becomes the spiral, flagellated band of the mature spermatozoid. In the region of the nucellar beak of the Ginkgo spermatid, the early MTOC is subtended by several folds of nucellar membrane (as evidenced in Fig. 5 by the presence of nucellar pores in the membrane profiles). Distally, there is a developing MLS; this proximal membrane body may act as an anchoring point for the growing MLS, but this is quite speculative. Although it is uncertain whether one can ascribe a function to these parallel membrane folds, it is possible that in some way they serve as an anchor to the proximal region of the MLS and that the point of origin of the MLS and these membranes was first established by the nucellar beak in contact with the spermatid's plasma membrane. At its distal end the growing MLS is associated with an MTOC, as seen in Fig. 4.

As can be seen in comparing the early transitional stages of Ginkgo blepharoplast dissolution in Figs. 1, 2 (and see fig. 10 in Gifford & Lin, 1975) with the blepharoplasts of Zamia and Microcycas in Fig. 7, those of cycads are larger but otherwise similar in organization. However, the blepharoplast of Zamia is unlike that of Ginkgo in that the latter retains its electron-opaque interior during its transition to the dispersal of probasal body clusters, whereas that of Zamia expands into a roughly spherical array of probasal bodies and there is no electron-opaque interior (Figs. 2, 8). However, it must be noted that we have no information regarding this stage in Microcycas.

[FIGURES 8-11 OMITTED]

In summary, the transition of blepharoplast to MLS in cycads and Ginkgo is similar, but the former lack participation of a nucellar beak in this phase of development. Also, these two examples differ in another aspect of development. As noted above, the posterior that is the first formed region of the Ginkgo MLS is subtended by several layers of parallel membranes (Fig. 6), whereas a comparable region in the MLS of Zamia exhibits an array of microtubules, which Norstog (1986) calls simply "the PMT" (posterior microtubules), suggesting that it has an anchoring function for the elongating MLS. Whether a similar function applies for the Ginkgo MLS is very speculative, but the structures in question may have analogous roles in spermatid development. Even so, we think they represent a marked and perhaps phylogenetically significant difference in spermatogenesis of both forms.

D. MORPHOLOGY OF THE SPERMATOZOID

In size and development, spermatozoids of Ginkgo and most cycads differ appreciably. The sperm of Ginkgo is about 86 [micro]m in diameter, whereas those of Zamia integrifolia measure 300 [micro]m in their greatest dimension and those of Z. chigua are huge--nearly 500 [micro]m in diameter (Norstog, 1977). Only Microcycas sperm are of about the same size as those of Ginkgo (Norstog, 1990). In addition to size, a great difference exists in the development of the flagellature between the sperm of Ginkgo and the sperm of cycads. With the exception of certain Microcycas sperm, those of cycads bear 5-7 gyres of flagella, whereas Ginkgo sperm bear fewer than 3 gyres.

Microcycas is a special case among cycads and Ginkgo in that it bears several pairs of sperm per microgametophyte. Up to 16 sperm have been observed in a pollen tube (Caldwell, 1907), and a dozen or so are not uncommon (Fig. 12). These are sequentially produced by divisions of a so-called sterile cell so as to form a linear array of spermatogenous cells, with the first-formed one farthest from the tip of the pollen tube and the most recently formed one at the tip. This arrangement also applies to the spermatozoids, but these are formed simultaneously instead of progressively (Fig. 12). Although all of the spermatogenous cells appear to be similarly developed, with blepharoplasts of the same size and having the same number of probasal bodies, the subsequently formed spermatids and spermatozoids vary greatly in their state of development. Those formed last and nearest the tip of the tube are spherical and bear only 1-2 gyres of flagella; the first-formed sperm, on the other hand, bear 5-6 gyres of flagella and are egg shaped (Fig. 12). As might be suspected, the first-formed sperm also swim much more vigorously than do the "lethargic" last-formed ones. Norstog speculates that a similar development in ancestral cycads, by eliminating slower and dysfunctional last-formed sperm, may have evolved in the direction of diplospermy characteristic of spermatophytes in general (it might be added that Chamberlain, in 1912, reported that, occasionally, four spermatozoids are present per pollen tube in Ceratozamia). Following this line of speculation, it is possible that ancestors of Ginkgo at one time produced fast-moving spermatozoids equipped with more gyres of flagella than are those of the present day.

[FIGURE 12 OMITTED]

V. Conclusions

Although both cycads and Ginkgo are products of ancient lineages of spermatophytes and are unique among extant seed plants in bearing motile sperm, the detailed structure of their respective motile cells argues against their having had any past close phylogenetic relationship. Rather, they seem to us to point to the ubiquity of primitive and, no doubt, widespread zooidogamy among ancient seed plants. Although the presence of motile male gametes is a general unspecialized condition (plesiomorphic) in vascular cryptograms, Ginkgo, and cycads among extant plants (Fig. 13), there are features in male gametophytes and spermatozoids that appear to be derived (apomorphic) in Ginkgo and others that appear to be derived within the cycads (Fig. 13). The presence of a nucellar beak in Ginkgo is found nowhere else in the green plants that have been investigated to date and, thus, is an autapomorphy for Ginkgo.

The presence of two male gametes (diplospermy) is a synapomorphy of seed plants (Fig. 13). This is further supported by the presence of two sperm in a pollen grain of the medullosan seed fern Pachytesta (Stewart, 1951), which is on the stem lineage of the extant seed plants (Nixon et al., 1994); i.e., it would be the branch between the Marattiales and the seed plants in Fig. 13. The singular presence of multiple sperm (multispermy) in Microcycas among extant seed plants is thus an autapomorphy of Microcycas within extant seed plants (Fig. 13). This conclusion is in contrast to that of Norstog (1995) and Norstog and Nicholls (1997), who considered the presence of multispermy in Microcycas to represent the retention of the ancestral vascular cryptogam and seed-plant condition. However, given the topology of the phylogenetic tree in Fig. 13, for this to be true diplospermy would have to evolved at least six times in extant seed plants. In contrast, the interpretation of a reversal to multispermy in Microcycas involves only two steps of character change. This interpretation does not reject the notion that multispermy is a plesiomorphic condition but rather simply interprets it as a single reversal in Microcycas instead of the multiple parallel development of diplospermy in other seed plants and other cycads. The same would be true even if multispermy were to be conclusively demonstrated for Ceratozamia. It is interesting to note that the presence of multispermy in vascular cryptograms and in Microcycas are the product of the production of multiple spermatogenous cells, as compared with the single spermatogenous cell produced in diplospermous seed plants. In fact, in all vascular plants each spermatogenous cell produces only two spermatids; thus it is the presence of several spermatogenous cells that leads to multispermy in Microcycas and the vascular cryptograms. Even so, we must point out that character analyses of gametophytes among cycads, to say nothing of those of other spermatophytes, are sparse. Gametophyte generations no doubt have their own evolutionary histories, and these may not mirror those of their equivalent sporophytes (Norstog, 1995; Philipson, 1991). In Microcycas, for example, both the microgametophyte and the megagametophyte are apparently primitive, for both are multigametic (up to 100 or more archegonia per megametophtye, though only a few are functional).

[FIGURE 13 OMITTED]

The massive sperm of the cycads appears to be an autapomorphy of the group. Generally speaking, cycads with diplospermy have sperm ranging from a minimum of 180 [micro]m in diameter for Cycas revoluta to 500 [micro]m for Zamia roezlii (Norstog & Nicholls, 1997). Microcycas is unusual in producing smaller spermatozoids, ranging from 60-80 [micro]m in diameter, similar to the size range for Ginkgo (ca. 60-85 [micro]m). What distinguishes, for a cycad, the small spermatozoids of Microcycas, as compared with those of Ginkgo, is that production of a linear series of spermatogenous cells in Microcycas, with each successive spermatogenous cell in turn producing ever-smaller pairs of spermatozoids. Thus the relatively small spermatozoids of Microcycas, as compared with other cycads, is result of process of diminution, as compared with the small size of spermatozoids of vascular cryptograms and other seed plants.

VI. Literature Cited

Bell, P, R. 1974. The origin of the multilayered structure in the spermatozoid of Pteridium aquilinum. Cytobiologie 8: 203-212.

Caldwell, O. W. 1907. Microcycas calocoma. Bot. Gaz. 44: 118-141.

Chamberlain, C. J. 1898. The homology of the blepharoplast. Bot. Gaz. 26: 431-435.

--. 1912. Morphology of Ceratozamia. Bot. Gaz. 53: 1-19.

Duckett, J. G. 1973. An ultrastructural study of the differentiation of the spermatozoid of Equisetum. J. Cell Sci. 12: 95-120.

Favre-Duchartre, M. 1956. Contribution/t l'etude de la reproduction chez le Ginkgo biloba. Rev. Cytol. Biol. Veg. 17: 1-217.

Friedman, W. E. 1987. Growth and development of the male gametophyte of Ginkgo biloba within the ovule (in vivo). Amer. J. Bot. 74: 1797-1815.

-- & E. M. Gifford. 1988. Division of the generative cell and late development in the male gametophyte of Ginkgo biloba. Amer. J. Bot. 75: 1434-1442.

-- & --. 1997. Male gametophyte of Ginkgo. Pp. 29-49 /n T. Hori (ed.), Ginkgo biloba, a global treasure: From biology to medicine. Springer-Verlag, Tokyo.

Gifford, E. M., Jr. & S. Larson. 1980. Developmental features of the spermatogenous cell in Ginkgo biloba. Amer. J. Bot. 67: 119-124.

-- & J. Lin. 1974. Ultrastructure de cellules spermatogenes de Ginkgo biloba, et particierment du blepharoplaste (mastigosome). Compte Rendus, Academie des Sciences (Paris) 278D: 2513-2515.

-- & --. 1975. Light microscope and ultrastructural studies of the male gametophyte in Ginkgo biloba: The spermatogenous cell. Amer. J. Bot. 62: 974-981.

Hepler, P. K. 1976. The blepharoplast of Marsilea: Its de novo formation and spindle association. J. Cell Sci. 21: 361-390.

Ikeno, S. & S. Hirase. 1897. Spermatozoids in gymnosperms. Ann. Bot. 11 : 344-345.

Larson, S. & E. M. Gifford. 1988. Division of the generative cell and late development of the male gametophyte of Ginkgo biloba. Amer. J. Bot. 75: 1434-1442.

Lee, C. L. 1955. Fertilization in Ginkgo biloba. Bot. Gaz. 117: 79-100.

Li, Y., F. H. Wang & R. B. Knox. 1989. Ultrastructural analysis of the flagellar apparatus of Ginkgo biloba. Protoplasma 149: 57-63.

Mizukami, I. & J. Gall. 1966. Centriole replication, II: Sperm formation in the fern, Marsilea, and the cycad, Zamia. J. Cell Biol. 29:97-111.

Nixon, K., W. Crepet, D. Stevenson & E. Friis. 1994. A reevaluation of seed plant phylogeny. Ann. Missouri Bot. Gard. 81: 484-533.

Norstog, K. 1974. Fine structure of the spermatozoid of Zamia: The vierergruppe. Amer. J. Bot. 61: 449-456.

--. 1975. The motility of cycad spermatozoids in relation to structure and function. Biol. J. Linn. Soc. 7 (suppl. 1): 135-142.

--. 1977. The spermatozoid of Zamia chigua. Bot. Gaz. 138: 409-412.

--. 1986. The blepharoplast of Zamia pumila L. Bot. Gaz. 147: 40-46.

--. 1990. The spermatozoid of Microcycas calocoma: Ultrastructure. Bot. Gaz. 151:275 284.

--. 1993. Spermatogenesis in Microcycas: Evolutionary significance of male gametes of seed plants. Pp. 270-278 in D. W. Stevenson & K. J. Norstog (eds.), The biology, structure, and systematics of the Cycadales: Proceedings of CYCAD 90, the Second International Conference on Cycad Biology. Palm & Cycad Societies of Australia Cycads, Milton, Queensland, Australia.

--. 1995. Spermatogenesis in Microcycas: Implications in the evolution of diplospermy in spermatophytes. Pp. 203-211 in D. D. Pant, D. D. Nautiyal, A. N. Bhatnagar, K. R. Surange, M. N. Bose & P. K. Khare (eds.), Proceedings of the International Conference on Global Environment and Diversification of Plants through Geological Time. South Asian Publishers, Allahabad, India.

-- & T. Nicholls. 1997. The biology of the cycads. Cornell Univ. Press, Ithaca, NY.

Philipson, W. R. 1991. A new approach to the origins of vascular plants. Bot. Jarb. Syst. 113: 443-460.

Stewart, W. N. 1951. A new Pachytesta from the Berryville locality of southeastern Illinois. Amer. Midl. Nat. 46: 716-742.

Turner, F. R. 1966. Changes in cellular organization during spermatogenesis. Ph.D. diss., University of Texas, Austin.

Webber, H. J. 1901. Spermatogenesis and fecundation of Zamia. U.S. Department of Agriculture, Bureau of Plant Industry, Bulletin No. 2.

KNUT J. NORSTOG

Missouri Botanical Garden 

Saint Louis, MO 63110, U.S.A.

ERNEST M. GIFFORD

Section of Plant Biology

Division of Biological Sciences

University of California

Davis, CA 95616, U.S.A.

AND

DENNIS WM. STEVENSON

New York Botanical Garden 

Bronx, NY 10458, U.S.A.