quantitative cytology of the sperm cells of

Planta (1987) 170: 446452 Planta �9 Springer-Verlag 1987

Quantitative cytology of the sperm cells of Brassica campestris and B. oleracea

Cameron A. McConchie 1 *, Scott D. Russell 2.*, Christian Dumas 3, Michael Tuohy 1 and R. Bruce Knox 1 1 Plant Cell Biology Research Center, University of Melbourne, Parkville, Victoria 3052, Australia, 2 Department of Botany and Microbiology, University of Oklahoma, Norman, OK 73019, USA, and 3 Universit6 C1. Bernard-Lyon 1, Biologie v6g6tale, F-69622 Villeurbanne Cedex, France

Abstract. Pollen grains of Brassica campestris L. var. acephala DC and B. oleracea L. were serially sectioned and examined using transmission electron microscopy to determine the three-dimensional organization of sperm cells within the micro- gametophyte and the quantity of membrane- bound organelles occurring within each cell. Sperm cells occur in pairs within each pollen grain, but are dimorphic, differing in size, morphology and mitochondrial content. The larger of the two sperm cells (Sv,) is distinguished by the presence of a blunt evagination, which in B. oleracea wraps around and lies within shallow furrows on the vegetative nucleus and in B. campestris can penetrate through internal enclaves of the vegetative nucleus. This sperm cell contains more mitochondria in both species than the second sperm cell (Sua). This latter cell is linked to the first by a common cell junction with the Sv,, but is not associated with the vegetative nucleus and lacks a cellular evagination. Such differences are indicative of a system of cytoplasmic heterospermy in which sperm cells possess significantly different quantities of mitochondria.

Key words: Brassica (sperm cell) - Cytoplasmic inheritance - Double fertilization - Mitochondrion - Pollen - Vegetative nucleus.

Introduction

In angiosperms, double fertilization results in the fusion of male gametes with both the egg and cen-

* Presen t address: Zellenlehre, UniversitM Heidelberg, Im Neuenheimer Feld 230, D-6900 Heidelberg, Federal Republic of Germany ** To whom correspondence should be addressed

Abbrev ia t ions ." mtDNA = mitochondrial DNA; S,~ = sperm cell unassociated with the vegetative nucleus; Svn sperm cell physically associated with the vegetative nucleus

tral cell, resulting in the formation of the embryo and the surrounding endosperm, respectively. The sperm cells, arising from a single mitotic division of the generative cell, have been regarded as identi- cal cells in both light (Maheshwari 1950, chpt. 5) and early electron-microscopic literature (Jensen and Fisher 1968; for review: see Russell 1986; McConchie and Knox 1986). With the exception of paternal plastid transmission, which has been reported for a minority of the angiosperms (Gill- ham 1978), the male cytoplasmic contribution to the offspring has largely been disregarded.

In a number of angiosperms, however, differences in sperm cells have been reported (for re- views: see Russell 1986; McConchie and Knox 1986). In many plants, physical associations occur between the vegetative nucleus and only one of the two sperm cells (Russell 1986), with the major exception being the grasses (Heslop-Harrison and Heslop-Harrison 1984; Mogensen and Rusche 1985). However, even in the grasses, sperm dimorphism has been found (Zea mays: McConchie et al. in press) and such a dimorphism may relate to preferential fertilization as demonstrated in pre- vious genetic studies of nuclear heterospermy in- volving disjunctional B-chromosomes in Zea (Ro- man 1948).

In Plumbago zeylanica, as far as we are aware the only dispermic plant for which quantification on organelles, cell, nuclear and cytoplasmic volumes is available, essentially only one sperm cell contains plastids and the other contains nearly five times as many mitochondria (Russell and Cass 1981; Russell 1984). The plastid-rich sperm cell preferentially fuses with the egg, while the mitochondrion-rich sperm cell enters the central cell (Russell 1985). All ultrastructural evidence indi- cates that male cytoplasmic organelles are viable when transmitted into female reproductive cells (Russell 1983).

C.A. McConchie et al. : Sperm of Brassica 447

In the present paper we report what, again, to the best of our information is the first quantification of organelles and of cell, nuclear and cytoplasmic volumes in two typical dispermic species, namely, Brassica campestris and B. oleracea. Bras- sica is more representative of angiosperms than Plumbago in possessing synergids, and in lacking plastids in the sperm. Cytoplasmic differences in the sperm cells of the two Brassica species are described and compared, and the potential for transmission of heritable organelles by sperm cells is evaluated.

Materials and methods

Plants of Brassica oleracea L. var. acephala DC and B. campestris L. were grown in a greenhouse from seed. Anthers containing mature pollen were collected near anthesis. Brassica oleracea anthers were fixed for 2-4 h in 3% glutaraldehyde in 0.1 M sodium-cacodylate buffer (pH 7.4) containing 1 mM CaC12 and 3% sucrose. They were then rinsed 12-18 h in several changes of buffer, fixed for 1 h in 1% buffered OsO,, briefly rinsed in buffer, dehydrated in a graded ethanol series, and embedded in epon (Dumas et at. 1985). Brassica campestris was fixed 4 h in 3% glutaraldehyde in 0.1 M K/Na-phosphate buffer, pH 6.8, and post-fixed I h in 1% buffered OsO4, dehydrated in a graded ethanol series, stained en bloc at 70% ethanol with a saturated solution of uranyl acetate, with tissue embedded in Spurr's low-viscosity resin (see McConchie et al. 1985). Sec- tions were stained with either uranyl acetate and lead citrate (UA-Pb) or periodic acid-thiocarbohydrazide-silver proteinate (PA-TCH-SP), according to methods described by Roland (1978).

All electron micrographs were taken using a Siemens 102 transmission electron microscope and printed at the same mag- nification for analysis. The volume and surface area of the sperm cells and nuclei were determined by measuring the area and circumference of each section using the graphics tablet of a Zeiss Videoplan and multiplying by the section thickness. This was estimated by gold interference color as being approx. 80 nm. Individual organelles were counted directly by compar- ing all of the electron micrographs in the series.

Three-dimensional reconstructions were prepared using prints of serial thin sections magnified to 18000-fold using either cardboard or cork gasket sheeting of 1.4 mm thickness (McConchie et al. 1985).

Results

General cytology. The mature pollen is tricellular in both Brassica species, with sperm cells formed within the pollen grain prior to anthesis. Each of the two sperm cells is bounded by a plasma membrane and together enclosed in the inner plasma membrane of the vegetative cell (Figs. 1-3). Both cells contain a normal complement of organelles including mitochondria, endoplasmic reticulum, ri- bosomes, microtubules, Golgi bodies, vesicles (Figs. I-3) and a prominent nucleus with a single nucleolus; however, both cells lack plastids

(McConchie et al. 1985; Dumas et al. 1985). Mito- chondria are small and ellipsoid with simple mor- phologies.

The two sperm cells of each pollen grain can be distinguished because of differences in size, morphology, and association with the vegetative nucleus (McConchie etal . 1985; Dumas etal . 1985). The larger of the two sperm cells (S~n) is characterized by the presence of a blunt evagination (Fig. 4), which in B. oleracea wraps around and lies within shallow furrows of the vegetative nucleus, and in B. campestris can penetrate through internal cavities in the vegetative nucleus (Fig. 4, arrow), sometimes re-emerging on the opposite side of the nucleus (Figs. 4-6, asterisks). The second sperm cell (S~) is linked to the first by a common cell junction, but is not associated with the vegetative nucleus. In both species, the junct ion between the two sperm cells consists of an elabo- rate pairing of finger-like evaginations held within a common periplasm (Figs. 4-6). In B. oleracea, unlike B. campestris, multiple protrusions of the sperm cells may occur, and in some cases the Su~ may wrap around and nearly reach the opposite end of the vegetative nucleus (Fig. 4; Dumas et al. 1985).

Quantitative cytology. The two sperm cells of Bras- sica campestris differ very significantly (P < 0.01) in overall volume (Table I), with the S~n an average of 19.4% larger than the Sua. The two sperm also differ significantly (P<0.05) in surface area (Ta- ble 1), with the surface area of the Svn 15.0% larger than that of the S~a. The sperm ceils of B. oleracea displayed similar patterns of dimorphism (Ta- ble 1), with the S w an average of 18.4% larger in volume (P<0.01) and an average of 41.0% greater in surface area (P<0.05) than the Su,. The differences in relative surface area between B. campestris and B. oleracea result from the occurrence of a greater number of slender evaginations in B. oleracea (Figs. 4-6; Dumas et al. 1985). Surface- area differences between individual sperm cells were also more prevalent, as evidenced by differences in standard errors of the mean (Table 1). A statistically significant linear correlation (P<0.01) was detected between volumes of the Svn and Sua (r=0.98) in B. oleracea, whereas in B. campestris such a correlation was not detected ( P > 0.05).

Differences in nuclear volume and surface area of the dimorphic sperm cells were not significant (P>0.05) in both species of Brassica studied. In B. campestris, the average nuclear volume of the Sv, was 13.1% larger than that of the S~a (Table 1). Nuclear surface differed by only 2.6%, with

448 C.A. McConchie et al.: Sperm of Brassica

Figs. 1-3. Transmission electron micrographs of thin sections of mature pollen of Brassiea campestris and B. oleracea, gv - Golgi vesicles; /d = lipid droplet; rn = mitochondrion; mt = microtubules; rer = rough endoplasmic reticulum; S.~ = sperm not associated with vegetative nucleus; Svn = sperm associated with vegetative nucleus; V N = vegetative nucleus Fig. 1. Vegetative nucleus and associated evaginations (arrowheads) of the Svn of B. oleracea. UA-Pb staining, x 12000; bar = 1 pm Fig. 2. Pair of sperm cells and vegetative nucleus of B. campestris. Arrowheads indicate evaginations of Svn within cytoplasmic enclaves of the vegetative nucleus. UA-Pb staining, x 11 500; bar = 1 pm Fig. 3. Association of Svn and vegetative nucleus of B. campestris. A microtubular array is present at point of emergence of the evagination. Parts of distal segments of the same evagination are indicated by arrowheads. PA-TCH-SP staining, x 11 300; bar = 1 pm

Figs. 4-6. Three-dimensional reconstructions of the male germ unit of Brassica carnpestris (left) and B. oleracea (right). Junctions between the paired sperm cells are indicated by dotted lines. Terminus of the S,n evagination of B. campestris is indicated by asterisk

Fig. 4. AxiM view of male germ units. Sv, evagination enters enclave of vegetative nucleus (indicated at white arrow), emerges, and terminates. Complementarity of small protrusions of the sperm cells of B. oleracea with the surface of the nuclear envelope indicated by small, black arrows. • 9300; bar = 2 gm


Fig. 5, Same reconstructions as above viewed from 45 ~ above horizontal. Interior of vegetative nucleus of B. campestris is shown in longisection (hinged at arrow) to illustrate the extent of the Svn evagination, x 7300; bar = 2 I-tin

Fig. 6. Same models each rotated by 180 ~ to illustrate the distal view of the sperm cells with respect to the vegetative nucleus. x 8800; b a r = 2 gm

450 C.A. McConchie et al, : Sperm of Brassica

Table 1, Cell volumes, surface areas and nuclear volumes of the dinaorphic sperm cells of B. campestris (n= 7) and B. oleracea (n= 5) as determined from serial reconstruction of sperm cell pairs. S~ =sperm physically associated with the vegetative nucleus. S~, = sperm not associated wkh the vegetative nucteus. ResuIts of paired t-test as ai~plied to aIzove I~arameters: n.s. = not signilScant; * = P < 0 . 0 5 ; ** = P<0,01

Parameter B. campestris B. oleracea

Sperm Sv. Sperm S~a Sperm Svn Sperm Su~

Cell volume (gm 3) Range 9.01 - 12.27 7.10-10,67 13.92-20.31 10.41 - 18.26 Mean 10.80+ 0.40 9.04+ 0.40** 17.53_+ 1.39 14.81+- 1.70"*

Cell surface (gin 2) Range 35.51 -- 49.02 30.96 - 40.14 46.04 - 77.05 32.60 - 66.22 Mean 40.78+- 1.76 35.46+- 1.51" 65.10+ 5.32 46.17_+ 6.04*

Nuclear volume (~tm 3) Range 2.39 -4 .58 2 .56- 3.10 3.97-7.04 3.91 - 7,46 Mean 3.20_+ 0.35 2.83 _+ 0.I 1 n.s, 5.77 • 0.57 5,56 +- 0.65 n,s.

Nuclear surface (gm z) Range 7 .97- 15.73 7 .43- ] 5.15 10.10- 20.88 l 1 .38- 2] .78 Mean 10.76_+ 1.05 10.49_+ ~.14n,s. 16.26+- 1,76 16.10_+ 1.7] n.s,

Cytoplasm volume (lain 3) Range 6.39-9.27 4.27 -7_77 8.79 - 13,60 6.50 - 12,46 Mean 7.60+-0.46 6.22_+0.42* 11.76! 0,93 9.25+- 1,18"

Mitochondria Range 10-- 34 4 - 1 1 lJ -- 15 6 - 14 Mean 23.43 _+ 3.34 6.43 +_ 1 09 ** 13.2 • 0.73 %8 _+ 1 80 *

the S~, nuclei having slightly larger area. Nuclear volume differences in sperm cells of B. oleracea were even less pronounced, as nuclear volume of the S w was, on the average, less than 4% larger than the S,, (Table 1). Nuclear surface of the Sv, was an average of less than 1% larger. Differences in volume between the two sperm cells of B. oleracea may therefore be entirely cytoplasmic (Ta- ble 1), without a significant contribution from the nucleus. This was also true, but to a lesser degree, for B. carnpestris.

Mitochondrial content was greater in the Sv, of both plants by very significant margins (Ta- ble 1). In B. campestris the number of mitochondria ranged from 4 to 34, with the Svn receiving an average of 23.4 mitochondria, and the Sua receiving only 27.4% of this amount, or an average of 6.4 mitochondria (Table 1). Similarly, in B. oleracea, the Sv. received an average of 13.2 mitochondria and the Sua, an average of 9.8 mitochondria (Table 1). Mitochondrial content of the Svn was more highly variable in B. campestris, amounting to a SE of + 3.34 (Table 1). In B. oleracea, the mitochondrial content of the Sua was more than twice as variable, even though the Sv, contained 34.6% more mitochondria (Table I). Akhough a significant positive correlation between sperm-cell volume and mitochondrial content was noted in B. oleracea (r =0.88), such a correlation was not

Table 2. Comparison of vegetative nucleus volume and surface area in B. campestris (n = 7) and B. oleracea (n= 5) as determined from three-dimensional reconstruction. Results of analysis of variance for unequal variances as applied to above parameters : n.s. = not significant; * = P < 0.05

Parameter B. campestris B. oleracea

Nuclear volume (gm 3) Range 21.48 - 35.25 25A8 - 56.04 Mean 25.92_+ ~.75 43.6~+- 5.70*

Nuclear surface (gm z) Range 42.70-75.40 45.16- 120,67 Mean 56.27_+ 4.43 90 .10_+ 13.51 n.s.

statistically significant in B. campestris, indicating greater variability in mitochondrial content compared to volume in this species.

The vegetative nucleus in B. campestris had an average volume of 25.9 ~tm 3, with a surface area of 56.3 gm z (Table 2), whereas in B. oleracea, the vegetative nucleus was 68% larger (average 43.6 gin3), with a 60% larger surface area (average 90.1 I, tm2) , Variance in these parameters was nearly three times larger in B. oleracea (Table 2), reflect- ing greater variabiIity in shape arid size of the vegetative nucleus. Statistical analysis showed a significant (P<0.01) correlation between the volume of the vegetative nucleus and the combined volume


of the sperm cells in B. oleracea @=0.96); however, there was no significant relationship between vegetative-nucleus volume and sperm volume in B. campestris.

Discussion

The sperm cells of Brassica campestris and B. oleracea both display marked, statistically significant to highly significant dimorphism in sperm volume, surface area, and particularly mitochondria content, and may be regarded as displaying cytoplasmic heterospermy (sensu Russell 1985). Although displaying a similar pattern of dimorphism, as might be expected of two closely related taxa, these species appear different in regard to the physical association of the sperm and the vegetative nucleus. In B. campestris, this association typically in- cludes a complete penetration of the vegetative nucleus by the cytoplasmic projection of the sperm cell (Svn), whereas in B. oleracea it is more similar to the relationship described in Plumbago (Russell 1984), where the projection is situated entirely on the surface of the vegetative nucleus. In B. oleracea, there are less extensive evaginations than in Plumbago, but extensions may also occur in the S,a, similar to those reported in Spinacia (Wilms and van Aelst 1983; Wilms et al. 1986).

The minimum number of mitochondria reported for the Su, of B. campestris, 4, and B. oleracea, 6 (Tables 1, 2), are the lowest reported in an angiosperm male gamete to date and indicate that these sperm cells are likely to be almost entirely physiologically dependent on the vegetative cytoplasm of the pollen tube. With a combined range of 4-14 mitochondria in the S,~ of both species, the possibility of male cytoplasmic inheritance is greatly reduced in this sperm cell. In the other sperm cell, the combined range was from 11 to 15 mitochondria in B. oleracea and from 10 to 34 in B. campestris, which are still low numbers. Assuming independent sorting-out of these organelles, as is believed to occur following cytoplasmic recombination (Birky 1983), the enriched mitochondrial environment of either the egg or the central cell would almost certainly result in non-ex- pression of transmitted male mitochondrial DNA (mtDNA). The situation with regard to transmission of paternal plastid D N A is even more straight- forward, as plastids are regarded to be the conven- tional reservoir of their characteristic DNA (Gill- ham 1978), and are apparently absent in the sperm cytoplasm of Brassica (McConchie et al. 1985; Du- mas et al. 1985). These data are in agreement with

genetic studies (Gillham 1978). The low number of potentially heritable organelles in Brassica is indicative of one means by which naturally occurring cytoplasmic hybridity in this species may be avoided. On a molecular level, the mitochondria of Brassica sperm may also contain an incomplete or reduced complement of m t D N A as a conse- quence of the reduced size of the sperm compared with somatic cells, a situation similar to that found in the small, ellipsoidal mitochondria of several other higher plants (Bendich and Gauriloff 1984). Such mechanisms as organellar diminution (see Mogensen and Rusche 1985) may act in further reducing the organellar complement during pas- sage in the pollent tube. The reported numbers may therefore represent a maximum potential contribution of mtDNA, with actual transmission being less.

As sperm m t D N A is apparently excluded in the offspring of flowering plants (Gillham 1978) the elimination of the paternal mitochondrial ge- nome could occur through several possible mechanisms: (i) unequal apport ionment of organelles to form generative and vegetative cells; (ii) unequal apport ionment of organelles to form two hetero- morphic sperm cells; (iii) cytoplasmic pinching-off of organelle-containing cytoplasm during matura- tion; (iv) cytoplasmic exclusion during gametic fusion with the egg; (v) selective non-incorporation or destruction of male m t D N A preventing its incorporation in the embryo chondriome. Our study on Brassica provides further evidence for the second of these mechanisms. If the mitochondrion- poor Sua fertilizes the egg, as in preferential fertilization of Plumbago (Russell 1983, 1985), the chances of incorporation of male mitochondria in the embryo are essentially nil.

Although evaluation of preferential fertilization in Brassica has not yet been completed, the occurrence of consistent size, shape, and mitochondrial differences in the sperm may represent a part of the outward attributes of such a system. The possibility exists for sophisticated mechanisms of gametic recognition and consistent patterns in sperm transmission. A fuller characterization of and greater insight into the male gametes of these two plants will require more unconventional meth- odology including isolating these gametes (Dumas et al. 1984; McConchie and Knox 1986).

We thank the Australian Department of Education (Special Research Centres Program) and National Science Foundation (grant PCM-8409151) for financial support, and Karen McCoy and Terryn Hough for skilled assistance with ultrastructural analyses and three-dimensional reconstructions.

452 C.A. McConchie et al. : Sperm of Brassica

References

Bendich, A.J., Gauriloff, L.P. (1984) Morphometric analysis of cucurbit mitochondria: the relationship between chondriome volume and DNA content. Protoplasma 119, 1-7

Birky, C.W. (1983) Relaxed cellular controls and organelle heredity. Science 222, 466-475

Dumas, C., Knox, R.B., Gaude, T. (1985) The spatial association of the sperm cells and vegetative nucleus in the pollen grain of Brassica. Protoplasma 124, 168-174

Dumas, C., Knox, R.B., McConchie, C.A., Russell, S.D. (1984) Emerging physiological concepts in fertilization. What's New in Plant Physiol. 15, 17-20

Gillham, N.W. (1978) Organelle heredity. Raven Press, New York

Heslop-Harrison, J., Heslop-Harrison, Y. (1984) The disposi- tion of gamete and vegetative-cell nuclei in the extending pollen tube of a grass species, Alopecurus pratensis L. Acta Bot. Neerl. 33, 131-134

Jensen, W.A., Fisher, D.B. (1968) Cotton embryogenesis: the sperm. Protoplasma 65, 277-286

Maheshwari, P. (1950) An introduction to the embryology of angiosperms. McGraw-Hill, New York

McConchie, C.A., Hough, T., Knox, R.B. (198) Ultrastruc- rural organization of the sperm cells of Zea mays. Proto- plasma (in press)

McConchie, C.A,, Jobson S., Knox, R,B. (1985) Computer- assisted reconstruction of the male germ unit in pollen of Brassica campestris. Protoplasma 127, 57-63

McConchie, C.A., Knox, R.B. (1986) The male germ unit and prospects for biotechnology. In: Pollen biotechnology and ecology, pp. 289-296, Mulcahy, D.L., Mulcahy, G.B., Otta- viano, E., eds. Springer, New York

Mogensen, H.L., Rusche, M.L. (1985) Quantitative analysis of barley sperm: occurrence and mechanism of cytoplasm and organelle reduction and the question of sperm dimorphism. Protoplasma 128, 1-13

Roland, J.C. (1978) General preparation and staining of thin sections. In: Electron microscopy and cytochemistry of plant cells, pp. 1-62, Hall, J., ed. Elsevier-North Holland, Amsterdam

Roman, H. (1948) Directed fertilization in maize. Proc. Natl. Aead. Sci. USA 34, 36-42

Russell, S.D. (1983) Fertilization in Plumbago zeylanica: gametic fusion and fate of the male cytoplasm. Am. J. Bot. 70, 416-434

Russell, S.D. (1984) Ultrastructure of the sperm of Plumbago zeylanica: 2. Quantitative cytology and three-dimensional reconstruction. Planta 162, 385-391

Russell, S.D. (1985) Preferential fertilization in Plumbago: ultrastructural evidence for gamete-level recognition in an angiosperm. Proc. Natl. Acad. Sci. USA 82, 6129 6132

Russell, S.D. (1986) Dimorphic sperm cells, cytoplasmic transmission, and preferential fertilization in Plumbago zeylaniea. In: The chondriome, pp. 69-116, Mantell, S.H., Chapman, G.P., Street, P.F.S., eds. Longman, London

Russell, S.D., Cass, D.D. (1981) Ultrastructure of the sperms of Plumbago zeylanica. 1. Cytology and association with the vegetative nucleus. Protoplasma 107, 85-107

Wilms, I-t.J., van Aelst, A.C. (1983) Ultrastructure of spinach sperm cells in mature pollen. In: Fertilization and embryogenesis in ovulated plants, pp. 105-112, Willemse, M.T.M., van Went, J.L., eds. VEDA, Bratislava

Wilms, H.J., Lefrink-ten Klooster, H.B., van Aelst, A.C. (1986) Isolation of spinach sperm cells: 1. Ultrastructure and three- dimensional construction in the mature pollen grain. In: Pollen biotechnology and ecology, pp. 307-312, Mulcahy, D.L., Mulcahy, G.B., Ottaviano, E., eds. Springer, New York

Received 28 September; accepted 4 December 1986

quantitative cytology of the sperm cells of

Documents