ER Morphology and the Changing Demands of the Cell Cycle

While ultrastructural methods are suitable for studying the details of ER structure in all cell types, the use of fluorochromes such as the vital stain 3,3/-dihexyloxacarbocyanine iodide, DiOC6(3) (Terasaki et al. 1984) and GFP technology (Haseloff et al. 1997) are superior in visualizing dynamic changes in ER organization in three dimensions. Although DiOC6 (3) and other fluorochromes appear to stain the great majority of different ER domains in plant cells, there is still no proof that all domains are equally labelled by the dyes or by GFP-tagged ER-targeted molecules.

ER Organization in Interphase Cells

Although the ER was first described by light microscopists, a more precise portrayal became possible only when Ledbetter and Porter (1963) introduced glutaraldehyde as a fixative for the electron microscopy of ultrathin sections. This technique, however, has a major disadvantage because it yields a rather limited three-dimensional view of cellular structures due to the minute thickness of the sections. Nevertheless, improvement in staining techniques gave a first glimpse into the 3-D pattern of the ER, and showed that it is an interconnected membrane system of tubules and flat sheets in remotely related differentiated interphase cells such as storage parenchyma cells of legume cotyledons, maize root cells, stamen hair cells, epidermal cells including guard cells, and moss caulonemata (Hepler et al. 1990). A distinct part of the ER was found in the cell cortex, largely present as a tubular network in close proximity to the plasma membrane (PM). In the vicinity of the nucleus it often exists as lamellar sheets covered with ribosomes. Lamellar sheets are not restricted to the nuclear region but may also occur embedded in the cortical tER network, close to mitochondria, plastids (Fig. 1b), or other cell compartments (Hawes et al. 1981; Hepler 1981; Galatis and Apostolakos 1977).

Electron microscopic observations have been confirmed by tetracycline fluorescence studies on onion bulb scale epidermal cells (Drawert and RüfferBock 1964), by DiOC6(3) staining (Quader and Schnepf 1986; Quader et al. 1987) (Fig. 2c), and by UV and VeDIC microscopy (Lichtscheidl and Url 1987; Allen and Brown 1988; Lichtscheidl and Weiss 1988). A polygonal meshwork of branching ER tubules with flat lamellar sheets fitted into that network interlace the cell cortex of higher plant cells and moss protonema cells (Hepler et al. 1990) (Fig. 2b,c). Since DiOC6(3) is a vital stain with little effect on the cellular activities at a low dose of exciting light, a spatial idea of ER patterning can be obtained by combining images from stacks of sections acquired in the z direction with the CLSM. This reveals the connection between the cortical ER network to ER elements like lamellar sheets located deeper in the cytoplasm in the vicinity of the nucleus or long tubular strands (Quader et al. 1987, 1989) (Fig. 2c). The long tubular strands seen in this way appear to be the form by which the ER membrane is displaced to distant cellular locations, like pulling on elastic. The velocity with which these ER tubules extend as long strands within the polygonal network range from about 0.5 to 8 ^m/s (Lichtscheidl and Url 1990).

The continuum of the ER membrane system becomes particularly evident at sites where ER tubules flow into or leave flat ER sheets located deeper in the cytoplasm, giving the impression of a canvas fixed by several ropes. The results obtained with the vital stain DiOC6(3) have been confirmed in living cells by attempts employing GFP technology (Boevink et al. 1996; Brandizzi et al. 2003) (Fig. 2a). Splicing the ER retrieval peptide K/HDEL to GFP, or tagging GFP to domains of ER resident proteins such as BiP (Lee et al. 2002), calreticulin (Brandizzi et al. 2003), or proteins associated with (calmodulin-regulated ATPase, Hong et al. 1999) or residing in the ER membrane (cal-nexin, Irons et al. 2003), have resulted in a similar dynamic tER network with polygons and lamellar sheets of different shape and size. Astonishingly, the polygonal ER network is also displayed in Arabidopsis by a GFP-tagged MT-associated (plus end tracking) protein EB1 (Mathur et al. 2003).

The cortical polygonal network is subject to considerable dynamic modifications as a consequence of changing physiological conditions (Lee and Chen 1988; Knebel et al. 1990; Hepler et al. 1990; Ridge et al. 1999). ER tubules may expand two-dimensionally to lamellar sheets, which predominantly emanate at sites where three or more tubules merge (cross) or where closely located tubules fuse to a flat sheet. Conversely, lamellar sheets may also disintegrate into tubules (Quader 1990). The sliding movement of ER tubules in a polygon may lead to its transformation, resulting in a new polygon or in its elimination (Fig. 3).

The cortical ER network is tightly associated with the PM at particular sites, named immobile fixed sites as indicated by centrifugation experiments

Endoplasmic Reticulum Morphology

Fig. 2 ER visualized by GFP technology (a) or by confocal microscopy after vital staining with the fluorochrome DiOC6(3) (b-d). a Cortical polygonal network visualized in a tobacco BY-2 cell by GFP technology with an epifluorescence microscope. Bar as Fig. 2c (courtesy of Ch. Ritzenthaler, Strasbourg). b Polygonal ER network in the cortex of protonemata cells of Funaria hygrometrica. c Various morphological ER domains observable in the nuclear region of onion bulb epidermis cells. The image was processed from 15 single sections taken in the z direction. The polygonal ER network and large lamellar sheets are located in the cell periphery. Deeper in the cell the nucleus is recognizable due to its brightly fluorescent NE. Lamellar sheets are located in close nearness on both sides of the NE, verifiable by the fluorescence intensity of the two small lamellar sheets, the brighter one residing adjacent to the NE side directed toward the cell interior which is longitudinally traversed by long tubular strands. Bright fluorescent spots: oval form—amyloplasts; more rounded form—mitochondria. d High magnification of an area traversed by a bundle of long tubular ER strands displaying the close linkage to the polygonal network. Bar as Fig. 2c

Fig. 2 ER visualized by GFP technology (a) or by confocal microscopy after vital staining with the fluorochrome DiOC6(3) (b-d). a Cortical polygonal network visualized in a tobacco BY-2 cell by GFP technology with an epifluorescence microscope. Bar as Fig. 2c (courtesy of Ch. Ritzenthaler, Strasbourg). b Polygonal ER network in the cortex of protonemata cells of Funaria hygrometrica. c Various morphological ER domains observable in the nuclear region of onion bulb epidermis cells. The image was processed from 15 single sections taken in the z direction. The polygonal ER network and large lamellar sheets are located in the cell periphery. Deeper in the cell the nucleus is recognizable due to its brightly fluorescent NE. Lamellar sheets are located in close nearness on both sides of the NE, verifiable by the fluorescence intensity of the two small lamellar sheets, the brighter one residing adjacent to the NE side directed toward the cell interior which is longitudinally traversed by long tubular strands. Bright fluorescent spots: oval form—amyloplasts; more rounded form—mitochondria. d High magnification of an area traversed by a bundle of long tubular ER strands displaying the close linkage to the polygonal network. Bar as Fig. 2c

(Quader et al. 1987; Liebe and Quader 1994), by combining stacks of images from the same plane of focus at short time intervals (Knebel et al. 1990), or by VeDIC observations (Lichtscheidl and Url 1990). The two membranes, however, never fuse. As regards the identity of these immobile fixed sites, the plasmodesmata of course come into question (see below and Oparka and Wright, this volume), but they cannot make up for all of them. Freeze fracture studies on actively secreting cells in pea root tips have revealed connections between the ER and strong PM indentation sites (Craig and Staehelin 1988).

Fig. 3 ER tubule motility within the polygonal network. Diagrammatic representation of the three major types of ER tubule movement leading to altered or disintegrated polygons: a polygon transformation by tubular sliding; b polygon disintegration by tubular sliding, and c polygon formation by tubule branching. Stars mark immobile fixed sites, arrows indicate direction of tubular motion

Fig. 3 ER tubule motility within the polygonal network. Diagrammatic representation of the three major types of ER tubule movement leading to altered or disintegrated polygons: a polygon transformation by tubular sliding; b polygon disintegration by tubular sliding, and c polygon formation by tubule branching. Stars mark immobile fixed sites, arrows indicate direction of tubular motion

Similar results were obtained in freeze-substituted Drosera epidermal cells (Lichtscheidl et al. 1990).

A particular situation occurs when neighbouring cells are interlinked by plasmodesmata. Ultrastructural studies have indicated that tER elements extend in a highly compressed or constricted form through this plasmodes-matal tunnel, thus forming an ER continuum between neighbouring cells (Hepler 1982). The sub-domain character of plasmodesmatal ER has been convincingly demonstrated by Grabski et al. (1993), who showed that a fluorescent diacylglycerol derivative locating predominantly to the ER and the NE can pass through plasmodesmata, but not fluorescent phospholipid analogues which locate predominantly to the PM.

ER tubules of primary plasmodesmata originate from portions of the ER extending through the area of the cell plate which become entrapped and compressed by phragmoplast vesicles during cell plate formation. (Hepler 1982; Hepler and Gunning 1998; see also Sect. 3.2). The participation of the ER in the formation of secondary plasmodesmata is also established (Lucas 1995), but little is known about the molecular processes involved in locating ER tubules to the site(s) of connection. Since the ER is closely associated with the PM, it is imaginable that the ER tubule and the PM together are pushed or pulled through the forming cell wall channel.

The tubular ER is, however, not only in close spatial association to the PM, but it is also often seen to locate adjacent to other compartments such as the tonoplast and mitochondria in Drosera epidermal cells (Lichtscheidl et al. 1990) or moss plastids (Galatis and Apostolakos 1977) (Fig. 1b). These "sites of nearness" have been proposed to represent domains of lipid exchange among organelles between which vesicle trafficking does not occur (Staehe-lin 1997), but have also been suggested to be involved in bulk lipid or Ca2+ transfer (Levine 2004). A special case occurs in maturing pollen grains where the ER of the vegetative cell is found in close association with the PM of the generative cell (Hess 1993; Luegmayr 1993). Endosymbiotic bacteria as well as multicelluar parasitic intruders are often wrapped in layers of ER, forming a shield which separates the intruding organism from the host cytoplasm (see references in Lichtscheidl and Hepler 1996).

A special ER domain is the nuclear envelope (NE) which separates the nuclear matrix and the cytoplasm during interphase. The NE is not always of distinct spherical shape but is characterized by invaginations and grooves or may even assume a flattened disc-like shape (Collings et al. 2000). It is comprised of outer and inner NE membranes which join at the nuclear pore complex (NPC). The outer NE is morphologically continuous with the ER network and shows functional conformity since it has the ability to synthesize proteins (Mattaj 2004), whereas the inner NE is functionally directed to the nucleoplasma and anchors the nuclear lamina and the chromosomes. Most of our knowledge regarding proteins residing in NE membranes, such as the nuclear lamina receptor of the inner NE, comes from animal cells. Little is known about lipids and proteins specifically located to the plant NE (Meier 2000). The continuity of the ER and the NE is marked by a gate-like contraction of the ER tubule close to the confluence (Staehelin 1997), which does not prevent protein distribution from and into the NE (Napier et al. 1992; Denecke et al. 1995; Boevink et al. 1996; Zachariadis et al. 2001; Pay et al. 2002; Irons et al. 2003). GFP-tagged RAN-GTPase activating protein marks the NE rim but also distributes in the cytoplasm of interphase cells (Rose et al. 2004). The NE also becomes labelled in Nicotiana benthamiana by GFP-labelled tobacco mosaic virus movement protein (Reichel and Beachy 1998). Irons et al. (2003) succeeded in labelling the NE by expressing GFP-tagged human lamin B receptor in tobacco BY-2 suspension cells and followed the distribution of this inner NE protein during mitosis. Since higher plant cells lack centrioles, the MT organization and function of the perinuclear area/outer NE with respect to MT is noteworthy in view of the mitotic rearrangement of the NE (Schmit 2002).

ER Rearrangement During Mitosis and Cytokinesis

Plant cells, as most other eukaryotic cells, are distinguished by an open cell division, i.e. characterized by the breakdown of the NE while chromosomes pile up at the equatorial plane, and subsequent NE restoration after successful chromosome segregation. Distinct changes in the distribution of the ER have recently been indicated in dividing cells of different taxa (Tables 1 and 2). Early electron microscopy observations of ER reorganization, which mainly concerned the cell periphery, the region of the dividing nucleus, and the succeeding mitotic apparatus (MA) (Hepler 1980; Hawes et al. 1981; Hepler and Wolniak 1984) (Fig. 4a,b), have been confirmed and extended by studies involving ER visualization by immunofluorescence techniques (Napier et al. 1992; Denecke et al. 1995; Zachariadis et al. 2003) or GFP technology (Nebenfuhr et al. 2000; Irons et al. 2003).

The first striking insights into changes of the ER and NE during mitosis were provided by Hepler (1980), who investigated the fate of the NE-ER complex by osmium tetroxide/potassium ferricyanide staining in dividing barley cells and, in a subsequent study (Hepler 1982), the association of the ER with the developing cell plate and primary plasmodesmata during cytokinesis. He showed the aggregation of NE and ER elements in prometaphase after NE breakdown, partially as fenestrated lamellae at the spindle pole region. Moreover, parts of the stained membranes apparently encased the MA or even invaded the spindle. Using GFP-tagged lamin B receptor it was shown that the inner NE in dividing mammalian cells does not ves-iculate during mitotic breakdown but equilibrates with the ER (Ellenberg et al. 1977), and that NE breakdown starts with a partial disassembly of the nuclear pore complex (Lenart et al. 2003). Using a similar construct, no other ER element but the NE is labelled in stably transformed tobacco BY-2 suspension cells during interphase (Irons et al. 2003). With the onset of metaphase, the fluorescently marked NE emerges at the pole region and to tubular structures around the MA, and becomes relocated to the NEs of the daughter cells during telophase/cytokinesis. The ER accumulating at the pole regions and at the MA rim obviously function as sites of NE retraction during mitosis.

But what is the situation during the stages preceding prometaphase and metaphase in which the mitotic players line up? The MT pattern during pre-prophase is characterized by the formation of a distinct narrow band in the cell cortex, the MT preprophase band (MT-PPB), at whose site the future cell plate will merge with the parental cell wall (Mineyuki 1999). The progressive breakdown of the MT-PPB at prophase is accompanied by the formation of the prophase spindle with MTs arranged along the nuclear rim progressing from the two pole regions towards the plane of division. In interphase cells the ER organization is similar in angiosperm and gymnosperm root cells,

Table 1 ER organization during mitosis and cell division: gymnosperms

Mitotic stage ER arrangement (higher organization)

Interphase All ER domains—peripheral polygonal network, lamellar flat sheets,

NE—form a continuous meshwork Preprophase ER-PPB; peripheral polygonal network still exists but less compact

Prophase Formation of a bipolar ER spindle constituting a sheath around the MA; NE breaks down

Pro-/ No ER aggregation at the pole regions; progressing NE breakdown;

metaphase transformation of major parts of the bipolar ER sheath into bundles of smooth K-NE-ER extending into metaphase spindle along kinetochore MTs and chromosome arms

Anaphase Shortening of the K-NE-ER tubules in accord with kinetochore MTs, gathering finally at the pole regions; parallel interzonal ER tubules develop between the two daughter chromosome groups Telophase/ Increase of ER elements at the pole region and of the number cytokinesis of interzonal ER tubules; gradual formation of barrel-shaped ER phragmoplast; NE restoration

Table 2 ER organization during mitosis and cell division: angiosperms and pteridophytes

Mitotic stage ER arrangement (loose organization)

Interphase All ER domains—peripheral polygonal network, lamellar flat sheets,

NE—form a continuous meshwork Preprophase No ER-PPB has hitherto been detected with certainty Prophase ER elements begin to accumulate at the pole regions; formation of a loose bipolar ER spindle which constitutes a sheath around the MA Pro-/ ER accumulation mainly at the pole regions; NE breaks down;

metaphase bundles of smooth NE-ER tubules (K-NE-ER) extending from the pole region into the metaphase spindle along kinetochore MTs and chromosome arms Anaphase Strongest ER accumulation at the pole regions;

shortening of the K-NE-ER tubules in accord with kinetochore MTs and progression of chromosome separation; ER tubules traverse the space between daughter chromosome groups Telophase/ Formation of an ER phragmoplast; ER elements are detected cytokinesis in the region of the forming cell plate and in pteridophyte leaf cells with the tER network and flat sheets in the cell periphery and deeper in the cytoplasm, as revealed by immunofluorescence studies, although a distinct difference is observable in these plant systems during preprophase (Zachariadis et al. 2003).

Fig. 4 ER organization during mitosis visualized by conventional electron microscopy. a ER gathering in the pole region of dividing cells of the moss Marchantia during early prometaphase (courtesy of P. Apostolakos, Athens); b as in (a), late prometaphase, indicating the association of prophase spindle microtubule with the ER (courtesy of P. Apostolakos, Athens)

Fig. 4 ER organization during mitosis visualized by conventional electron microscopy. a ER gathering in the pole region of dividing cells of the moss Marchantia during early prometaphase (courtesy of P. Apostolakos, Athens); b as in (a), late prometaphase, indicating the association of prophase spindle microtubule with the ER (courtesy of P. Apostolakos, Athens)

In dividing root-tip cells of Pinus, the ER pattern at preprophase and prophase follows more or less that of microtubules forming a distinct bandlike structure of parallel ER tubules succeeded by a tER prophase spindle. The three-dimensional reconstruction of a series of CLSM sections showed unequivocally that the observed band of ER tubules constitutes a ring underneath the PM, a tER-PPB, which in early preprophase is still loosely arranged and apparently becomes more compact during the transition to early prophase, following the maturation pattern of the MT-PPB (Fig. 5a) (Zachari-adis et al. 2001, 2003). ER tubules have also been noticed among PPB MTs by electron microscopy (Mineyuki 1999). Of note is that the ER network traversing the cell cortex in interphase cells persists even after the formation of the tER-PPB, whereas all the MTs in the cell periphery except those of the MT-PPB disappear during PPB formation (Zachariadis et al. 2003). The function of the temporary tER-PPB is unknown, but it may play a role in Ca2+ sequestration/release which is important in regulating MT-PPB polymeriza-tion/depolymerization.

A tER-PPB could be visualized with certainty only in gymnosperms. It was not detectable in dividing leaf cells of the pteridophyte A. nidulans, and only sporadically in dividing angiosperm root-tip cells exposed to stress (Zachari-adis 2003, unpublished observations).

In dividing Pinus root cells, from prophase through metaphase and anaphase, the ER distribution closely resembles that of the microtubules as it forms a tER prophase spindle (Fig. 5a). It becomes transformed into a bipolar metaphase spindle characterized by relatively imprecise pole regions and bundles of ER tubules that span the area between the poles and the

5a

tj.î'

10 ym

a'

b

il

Ç 10 yn

I

d'

Fig. 5 ER organization during mitosis visualized by immunofluorescence techniques employing an antibody which recognizes the ER retention signal. Chromosomes were stained with propidium iodide. Note that the cortical ER network is present throughout mitosis/cytokinesis. a Pinus brutia root-tip cell, transition preprophase to prophase. The ER-PPB, the forming bipolar prophase ER spindle, and the arrangement of the chromosomes in the same cell. b Bundles of kinetochore ER tubules with broad pole regions during metaphase of a P. brutia root-tip cell and chromosome pattern. ER tubules traversing the interzonal area. Note the fir-tree character of the kinetochore ER bundles. c Triticum turgidum mid-anaphase root-tip cell with marked concentration of ER elements at the poles (mostly K-ERs) and ER tubules traversing the space between the two daughter chromosome groups. d ER phragmoplast of a dividing Pinus root-tip cell and chromosome arrangement

Fig. 5 ER organization during mitosis visualized by immunofluorescence techniques employing an antibody which recognizes the ER retention signal. Chromosomes were stained with propidium iodide. Note that the cortical ER network is present throughout mitosis/cytokinesis. a Pinus brutia root-tip cell, transition preprophase to prophase. The ER-PPB, the forming bipolar prophase ER spindle, and the arrangement of the chromosomes in the same cell. b Bundles of kinetochore ER tubules with broad pole regions during metaphase of a P. brutia root-tip cell and chromosome pattern. ER tubules traversing the interzonal area. Note the fir-tree character of the kinetochore ER bundles. c Triticum turgidum mid-anaphase root-tip cell with marked concentration of ER elements at the poles (mostly K-ERs) and ER tubules traversing the space between the two daughter chromosome groups. d ER phragmoplast of a dividing Pinus root-tip cell and chromosome arrangement chromosomes' kinetochores (K-tER), probably following the kinetochore MT bundles (Fig. 5b). These well-defined K-tER bundles gradually shorten during anaphase, while a system of interzonal ER tubules is formed between the two separating chromosome groups. The ER concentrates at the spindle poles partly due to the retraction of the K-tER bundles (Zachariadis et al. 2003). An efficient ER reorganization occurs in the transition from late anaphase through telophase and cytokinesis. There are few ER elements recognizable at the phragmoplast-forming site (Segui-Simarro et al. 2004), although the interzonal ER tubules appear to multiply in number and traverse the space between the separated chromosome groups as loose bundles. Progressively, the ER bundles assume the form of a barrel-shaped young ER phragmo-plast which converts, after shortening of the ER bundles, into a structure resembling the typical MT phragmoplast (Fig. 5d) (Hepler and Gunning 1998; Zachariadis et al. 2003).

The ER pattern in dividing Triticum root-tip cells differs from that of dividing Pinus root-tip cells, not only in the uncertainty regarding the formation of a tER-PPB, but also in the strong accumulation of tER elements at the spindle poles (Hepler 1980; Hawes et al. 1981) and in the formation of a less distinct interzonal tER array (Fig. 5c). Similar differences apply to the redistribution of the ER in dividing Asplenium leaf cells (Zachariadis et al. 2003). The ER pattern observed in Pinus during cell division shows a higher degree of organization than in the other species. This is probably due to the molecular difference between the MTs of gymnosperm and angiosperm meristematic root cells, since the MTs in Pinus have acetylated tubulin and MTs formed by this tubulin derivative are known to have a relatively longer half-life (Gilmer et al. 1999). The extended turnover of the acetylated MTs might provide a more stable framework for the organization of the ER. In contrast, it may be difficult to detect a tER-PPB in dividing angiosperm cells because of the higher turnover of the PPB MTs.

Zachariadis and co-authors (2003) have suggested from inhibitor experiments that the control of ER organization may switch in dividing Pinus root cells from AF dependence during interphase to MTs during mitosis and cytokinesis, because in the presence of microtubule inhibitors like oryzalin the formation of the tER-PPB, tER metaphase spindle, and tER phragmoplast is prevented. The actomyosin system shown to guide the ER pattern in interphase cells is apparently not involved in the shaping of the tER-PPB in dividing Pinus root cells, although AFs also form a PPB-like structure in these cells (Zachariadis et al. 2001, 2003). MT-dependent ER reorganization during mitosis and cytokinesis correlates with the fact that vesicles are steered along phragmo-plast MTs (Verma and Gu 1996). Mathur and co-workers (2003) reported a new EB1-like protein, AtEB1, in Arabidopsis that co-localizes to the growing MT plus ends and to motile endomembrane networks including the ER. Moreover, an integral membrane protein of the reticular sub-domain of the rough ER, p63, has been shown to bind MTs in vivo and in vitro in Xenopus (Klopfenstein et al. 1998). As yet no homologues of this protein have been found in plants.

Was this article helpful?

0 0

Post a comment