J. Cell Sci. Suppl. 2, 129-141 (1985)Printed in Great Britain © The Company of Biologists Limited 1985 129

CYTOLOGICAL AND BIOCHEMICAL REQUIREMENTS FOR THE ESTABLISHMENT OF A POLAR CELL

RALPH S. Q U A T R A N O * , L A W R E N C E R. G R I F F I N G , V E R O N I C AH U B E R - W A L C H L I a n d R. S C O T T D O U B E TDepartment o f Botany and Plant Pathology, Oregon State University, Corvallis,Oregon 97331-2902, U.SA.

SU M M A RY

Our research is aimed at understanding the biochemical and cytological basis of cell polarity in zygotes of the brown alga, Fucus distichus L. Powell. One manifestation of this polar cell is the localization of a sulphated fucan polysaccharide (F2) in only one region of the zygote cell surface, the rhizoid cell wall. The focus of this paper is centered around the mechanism responsible for the directional transport of Golgi vesicles containing Fz and the biochemical properties of F2 that might specify its localized fate.

Recent findings indicate that the various sulphated polysaccharides in the brown algae are com­plexes resulting from linkages of two basic polymers: an a-(l — * 2 )-linked fucan that contains high levels of ester sulphate (F3 ), and a uronic acid-rich polymer (Fj). The fucan complex F2 , which is localized in the rhizoid wall, is composed of a fucan sulphate core (F3) to which uronic acid polymers (similar to Fi) are attached. Our results, using a purified endoguluronate lyase, indicate that guluronate bridges link these subunits of F2 . The carbon backbone of F2 is not synthesized de novo after fertilization. However, F2 is sulphated, and possibly assembled, beginning 10 h after fertiliza­tion, after which it is locally inserted into the rhizoid wall, and held in the wall structure only by calcium ionic bonds. Although sulphation is required for localization of F2, it is not known if the uronic acid side-chains are also assembled at the time of sulphation, and/or required for localization. The fact that F3 (F2 without the side-chains) is secreted uniformly into the zygote wall suggests that the uronic acid chains of F2 may play a critical role in its localization.

A sulphated F2 alone is not sufficient for its localization since in the presence of cytochalasin, vesicles containing F2 are not transported to the rhizoid. Recent studies point to a central role for a cytoskeletal element, possibly microfilaments, in the directional transport of these vesicles. We have used the techniques of isoelectric focusing and electrophoretic mobility to study surface charge of these Golgi vesicles to determine if charge might be one factor that specifies their localization. Vesicles that contain the sulphated fucan F3 are secreted randomly and have the same surface charge as those containing F2

that are directionallytransported. However, there is no stable endogenous electrical current at the time when F3 vesicles are randomly secreted, whereas a current is detectable when F2 vesicles are localized. The same vesicle population without a sulphated F2 is not localized and has a significantly lower negative surface charge than the vesicles containing F2 or F3 . Hence, surface charge is related to sulphation and ability to localize and is consistent with a self-electrophoretic mechanism of localiza­tion. The relationships between the sulphation, assembly and packaging of F2 into the Golgi vesicles, as well as the surface properties of the localized vesicles (other than charge) and a cytoskeletal com­ponent, must now be investigated. A discussion of our previous model for the establishment of a polar cell is presented along with various modifications in the light of more recent evidence.

IN T R O D U C T IO N

A central question facing cell and developmental biologists is, how does an ex­tracellular gradient cause the localization of subcellular components and surface

* Author for correspondence.

130 R. S. Quatrano and othersparticles leading to the establishment of a polar cell? Zygotes and embryos of the brown algae have been used for about 100 years as a model system to study the development of cell polarity (cf. Quatrano, 1978; Evans, Callow & Callow, 1982). Spherical zygotes of the Fucales respond to unilateral light and numerous other gradients by forming a localized outgrowth or rhizoid, which is oriented with respect to the imposed gradient (cf. Jaffe, 1968). In addition, the structural and biochemical polarity exhibited in the zygote at the time of rhizoid formation arises from a pre­viously homogeneous egg cytoplasm at a site that is determined by an external gradient (Quatrano, Brawley & Hogsett, 1979). All of these events and processes occur in a synchronously developing single cell system under completely defined conditions (Quatrano, 1980). Our goal is to understand the cytological and biochemi­cal basis for establishment of this cell polarity in the Fucus zygote.

The morphological manifestation of polarity is the initiation of rhizoid growth. This is preceded by three events: orientation of a labile polar cixis by an external gradient, the irreversible determination or fixation of this axis, and finally, the sub­cellular localization of vesicles and polysaccharides at the fixed site to support rhizoid tip growth. These processes can be separated experimentally from each other so that a population of zygotes can be used to study the biochemical and physiological events associated with axis orientation, axis fixation and site-specific localization (Quatrano, 1973).

The first sign of asymmetry in the zygote occurs during the time when the axis is labile and subject to reorientation by the same or different gradients. An endogenous electrical current begins to flow through the zygote with the positive pole and inward current localized at the presumptive site of rhizoid initiation (J affe & Nuccitelli, 1977). At least part of this current is carried by calcium (Robinson & Jaffe, 1975). If the orienting gradient, e.g. light, is reversed at this time, Nuccitelli (1978) showed, using a vibrating probe, that the spatial current pattern changed with an inward current present on the dark side, i.e. the new site of rhizoid formation. This change occurred within 40 min of the light reversal. As the time of axis fixation approaches, the inward current regions of the plasma membrane are concentrat ed at the predeter­mined rhizoid site. The basis for fixation is not known but a role for cytoskeletal elements is suggested by the result that axis fixation is prevented in the presence of cytochalasin (Quatrano, 1973). With the axis fixed, the zygote exhibits an endogenous electrical potential gradient, which has been postulated to localise molecules/part­icles to the appropriate poles of the zygote by ‘self-electrophoresis’ (Jaffe & Nuccitelli, 1977).

Between the time of axis fixation and rhizoid growth, intracellular components are transported to and accumulate at the cytoplasmic site of the rhizoid (Quatrano et al. 1979). One particular fucan polysaccharide is sulphated throughout the cytoplasm at this stage, transported in Golgi vesicles to the rhizoid site and insert ed into the rhizoid wall (Brawley & Quatrano, 1979; Evans et al. 1982). The focus of this paper is centered around the mechanism(s) responsible for the directional transport of these Golgi vesicles and the biochemical properties of the sulphated fucan that might specify its localized fate.

Requirements for cell polarity 131

C H A R A C T E R IZ A T IO N OF T H E S U L P H A T E D P O L Y S A C C H A R ID E S

Fucans from whole plantsBrown algae were known to possess an array of sulphated polysaccharides that

contained fucose as the predominant sugar. Sulphate was found attached to fucose by an ester linkage only. Mian & Percival (1973) concluded that there was a spectrum of fucose polymers, from high-uronic acid/xylose and low-sulphate species to a relative­ly homogeneous fucan sulphate. More recently, Medcalf & Larsen (1977a,b) found that 90 % of the total fucose-containing polymers were extracted with acid from whole Fucus vesiculosus plants. Two components were resolved from this extract; an ascophyllan-like component (ALC), which represented about 25% , and a fucan complex (F C ), which comprised the remainder. The ALC was a xylofucoglucuronan that contained about 22% uronic acid, 4% sulphate and 18% fucose. The main fucan polymer, FC, was composed of 6 % uronic acid, 25 % sulphate and about 50 % fucose. Both ALC and FC each contained 5 % protein, which apparently was linked to the polysaccharide by serine and threonine rather than through hydroxyproline. No polymer was found that contained only fucose and sulphate.

Medcalf & Larsen (1977a,b) also reported that when FC was hydrolysed by mild acid (20mM-HCl, 80 °C, 60min), the original band of FC on a cellulose acetate electrophoretogram disappeared and two new ones appeared: a trace of a slower fraction with a mobility equal to ALC, and a major band with a mobility faster than the original FC, approaching that of a pure fucan sulphate. Two similar products of hydrolysis from the brown alga Ascophyllum nodosum were also analysed by Medcalf & Larsen. They showed that the electrophoretically slower component (ALC) was enriched in protein (7%) and uronic acids (24%), and depleted in fucose (11%), while the electrophoretically faster component was high in fucose (42%) and had only traces of protein (2%) and uronic acids (4%). Generation of the two hydrolysis products could not be mimicked by proteolytic treatments, which indicated that the protein fraction was not a bridge but linked directly to the ALC. Various chemical treatments indicated that the FC was composed of a fucan backbone to which various numbers of ALCs were attached by an acid-labile linkage. The ALC components of the FC appeared to be composed of auronic acidbackbone, primarily mannuronic acid, and long fucose-containing side-chains that are low in sulphate. Some evidence was presen­ted that indicated that the ALC in the complex was different from free ALC.

Fucans from cell walls of embryosWe have recently obtained similar findings using purified cell walls from 24-h-old,

two-celledF. distichus embryos (Doubet, 1983). The ALC and FC components were detected in the acid-soluble fraction and will be referred to as Fi and F 2, respectively. An electrophoretically faster component (F3) was detected upon mild acid hydrolysis of cell walls, similar to the results of Medcalf & Larsen (19776) when F2 was hydrolysed. However, the electrophoretically slower component of F2 from the cell walls was apparently masked by the free Fi already present in the wall preparation.

132 R. S. Quatrano and othersWhen cell walls were treated with a specific endoguluronate lyase isolated from a marine bacterium (Doubet & Quatrano, 1984), an electrophoretic pattern was generated that was identical to that produced by mild acid hydrolysis (Fig. 1). This suggested that the fucan complex F 2 was composed of F 3 to which numerous A LCs were linked by a segment containing guluronic acids. This stretch of guluronic acids (G-block) probably corresponded to the acid-labile linkage described by Medcalf & Larsen (1977a). What appeared to be emerging from these studies was the presence of two primary fucan polymers, the ALC/Fi, and, a relatively homogeneous fucan sulphate, F 3 . We propose here that any number of fucan complexes (F 2) can be generated when A LCs are linked in various proportions and configurations to the F 3 backbone by G-blocks. T he model would be consistent with the chemical and enzymic hydrolysis data of ours and those of Medcalf & Larsen (1977a,b), as well as the chemical and fractionation studies of Mian & Percival (1973).

L O C A L I Z A T I O N O F T H E S U L P H A T E D P O L Y S A C C H A R I D E S I N Z Y G O T E

C E L L W A L L S

In order to determine if F i, F 2 and F 3 exist as such in the native cell wall, and to understand how these polymers were integrated into the wall, we treated purified zygote cell walls with E G T A and the purified endoguluronate lyase. F 2 was released intact from the cell wall only by treatment with EG T A , while Fi was removed intact by E G T A or by the lyase. F 3 could only be isolated from the wall structure by treatment with the lyase and was present even in the absence of F 2 (Fig. 1). These results suggest that F i, F 2 , F 3 exist as unique polymers in the native cell walls, and, that Fi is held in the wall structure by Ca2+ bridges via a G-block structure (similar to the gel-forming complex of the polyuronic acid, alginate; see Grasdalen,

No + E G T A + E G T A + L y asetreatm ent 6-5 h 24 h 6-5 or 24 h

0 - — him --------------- --------------- 1 1

0-2 -

' 0 - 4 - ■

R f2 0-6 - " 1 — - FI

0-8 -

1 1 - F31-2 -

Fig. 1. Tracings from stained cellulose acetate electrophoretograms of purified cell walls from 6-5 and 24h-old F. distichus embryos after treatment with E G T A and en­doguluronate lyase. Notice the presence of F 2 only in EG TA-treated 24 h walls. F 3 was detected only after lyase treatment of walls from 6-5 and 24 h embryos. Since F 2 was not found in 6'5 h walls, the presence of F 3 indicated that it was directly attached to the wall structure by G-block.

Requirements for cell polarity 133Larsen & Smidsrod, 1981), F2 by Ca2+ -bonding probably not involving G-block, and F 3 only by G-block directly to the wall structure or as a component of F2 (Doubet,1983).

Analysis of Fucus zygote cell walls isolated at different times after fertilization indicated that by 6 h after fertilization, Fi and F 3 are structurally integrated into the wall while F2 was not found until just before the emergence of the rhizoid, i.e. at about 14h (Fig. 1). Fi was newly synthesized after fertilization, whereas the carbon back­bones of F 2 and F 3 were either performed in the egg or synthesized from a carbon precursor pool different from carbonate fixation from seawater (Doubet, 1983). The sulphation of Fi and F 3 occurred during early stages of cell wall assembly i.e. before 6 h, whereas sulphation of F2 occurred after 10 h, just before rhizoid formation (Quatrano & Crayton, 1973; Hogsett & Quatrano, 1975, 1978). Sulphation of all fucan polymers occurs in the Golgi (cf. Evans et al. 1982) but the subcellular distribution of F 1/F3 and F2 was different. Fi and F3 were uniformly distributed throughout the zygote wall (cf. Quatrano, 1979). Autoradiographic evidence using 14 h zygotes pulsed for 5 min with 35S 0 4 , showed that sulphation of F 2 occurred in vesicles randomly located in the cytoplasm. After a ‘chase’, however, F2-containing vesicles were directionally transported to the emerging rhizoid portion of the zygote and F2 deposited only in the rhizoid wall (Brawley & Quatrano, 1979). In addition to these autoradiographic and other cytochemical staining data, there was ultrastructural (Vreugdenhill, Dijkstra & Libbenga, 1976) and biochemical evidence for a unique wall layer, i.e. F2, localized only on the surface of the rhizoid (cf. Quatrano et al. 1979). These characteristics of Fi, F 2 and F 3 are summerized in Table 1. What are the requirements for: ( 1 ) the directional transport of vesicles containing F2 to the rhizoid, and (2 ) the localized insertion of F2 into the rhizoid cell wall of the zygote?

R E Q U IR E M E N T S FO R L O C A L IZ A T IO N OF F 2

SulphationWhen zygotes were pulsed with 35S0 4 for 60 min at various times after fertilization,

the enzymic sulphation of F2 was initiated at 10 h. The pattern of sulphate accumula­tion into F2 was not due to changes in the pool size or permeability of the zygotes to sulphate, and, was dependent upon synthesis of new protein. A considerable amount of data suggested that this sulphation occurred on a pre-existing fucan rather than beingde novo synthesis of thqentire F polymer (Quatrano & Crayton, 1973; Quatrano et al. 1979). Consistent with the data would be the possibility of a pre-existing ALC and F 3 becoming sulphated before, during or after their assembly into F 2. The final assembly of F2 could therefore involve both sulphation and other modifications, such as the attachment of ALC to the fucan backbone, i.e. F3 (Fig. 2).

Although the time of sulphation coincides with the initial stages of F 2 localization, the formation of the rhizoid was not dependent upon this sulphation. Clayton, Wilson & Quatrano (1974) found that zygotes grown in seawater that lacked sulphate but contained methionine, formed rhizoids and two-celled embryos (Met-embryos).

134 R. S. Quatrano and others

Table 1. Comparison o f the properties o f the sulphated fucan polysaccharides F ItF 2 and F j from Fucus zygotes

F , (A LC ) f 2 f 3

Stabilized in the wall by 4 h with Ca2+ via G-blocks

Synthesized de novo and little or no sulphation detected

High in uronic acids and protein, equal amounts of xylose and fucose, and low in sulphate

Uniformly distributed in zygote wall

Stabilized in the wall after l Ohby Ca2+

Not synthesized de novo but highly sulphated before incorporation into cell wall

Composed of F 3 backbone and Fi/A L C side-chains, linked by G-block

Localized in rhizoid part of zygote wall

Stabilized in the wall after 4 h by G-blocks

Not synthesized de novo but highly sulphated before incorporation into cell wall

High in fucose and sulphate with traces of xylose, uronic acids and protein

Uniformly distributed in zygote wall

(0)

(3)' i ' . ' i ' i ' i ' i ' i '

i i i i i i i i I I I I I I T F,

X Xx (2) x x ' xX X

I X| I I I I x| | | I I I X I 1 1 1 'X

X XX X

X XX X

( 1)X X X X X X X X X X X X X X X X X X X X X X X X

Fig. 2. Diagram of the possible pathways by which the two major fucan polymers, the a - ( l —> 2)-linked fucan (0) and the A LC or Fi (1), were secreted into the wall without change (pathway C), or modified by sulphation (pathway A) and/or assembly (pathway B ). These gave rise to the sulphated fucans Fi (1), F 2 (2) and F 3 (3). Four hours after fertilization, pathways A and C were operating while pathway B, with possible contributions from pathways A and C, was initiated after 10 h. Evidence suggested that (0 ) is not synthesized de novo and was either preformed in the egg or resulted from carbon rearrangements of glucose released from the breakdown of the egg storage polymer laminarin. At least a portion of (1) was newly synthesized after fertilization. Sulphation (lllllll) definitely occurred after fertilization and possibly the attachment of the uronic acid side-chains ( X X X X X ) as well. Fj and F 3 were randomly distributed in the cell wall whereas F 2 was localized in the rhizoid. ( -------)a - ( 1 —> 2 ) -fucan polymer; (lllllll) ester sul­phate; ( X X X X X ) uronic acid-enriched polymer.

Since the endogenous electrical current has always been associated with rhizoid formation (Jaffe & Nuccitelli, 1977) we will assume that Met-embryos have a normal spatial current pattern. However, Met-embryos neither adhered to the substratum nor exhibited the cytrochemical localization of F 2 in the rhizoid cytoplasm or wall. Is

sulphation then specifically required for F2 to become localized ? If the sulphating sites and F 2 were localized in the absence of sulphation we should be able to detect this by short pulses of 35S and a specific stain for the unsulphated F2. Using fluoresent- labelled ricin as a specific cytological stain for F2, we showed that when zygotes were grown in the presence of SO4, a very intense fluorescence was observed at the rhizoid tip. If the localized F2 was then chemically desulphated the same localization of the stain was observed, indicating that if an unsulphated F2 was localized, it too would be detected by the stain (unlike the metachromatic stain Toluidine Blue O, TBO ). However, no localization of the ricin was observed in Met-embryos (Hogsett & Quatrano, 1978). Also, two-celled embryos grown with SO4 exhibited localized sulphating sites when pulsed for 5 min with 35S0 4 , whereas Met-embryos at the same developmental stage showed randomly distributed sites of sulphation (Brawley & Quatrano, 1979). Hence, the cytochemical and autoradiographic data were both consistent with the conclusion that enzymic sulphation of F2 was one requirement for its localized accumulation in the rhizoid wall. Can sulphation alone account for localization of F2?

Fz assemblyIt is possible that the ALC part of F 2, containing about 7% protein, may play a key

role in localization. The fucan backbone of F2, i.e. F 3, was itself preformed, sulphated and uniformly inserted within the zygote wall during early development, hours before sulphation of F2 and rhizoid formation (Hogsett & Quatrano, 1975). There is no evidence for F 3 being directionally transported to the rhizoid. The sulphation of F 3

alone, then, was not sufficient for directional transport and localized insertion into the wall. Perhaps it must be modified further by attachment of ALC side-chains in order for localization to occur. It is also important to point out that we do not know if zygotes grown in the absence of SO4 were inhibited in both the assembly of F2 (linking of ALC side-chains to F 3 by G-block) and sulphation. At present there is no way of separating unsulphated ALC, F 2 and F 3 from each other, and nothing is known about the synthetic steps involved in F2 assembly. It appeared then that sulphation and possibly a post-synthetic modification of F3 to give a sulphated F2 at 10 h were the biochemical modifications of F2 necessary for its localization (Fig. 2).

Cytochalasin-sensitive processDirected transport of vesicles containing F2, and F 2 localization in the rhizoid wall,

also required a cytochalasin (CH)*-sensitive process (Quatrano, 1973). Although microfilaments have not been demonstrated conclusively in Fucus zygotes, Brawley & Robinson (1985) recently demonstrated a change in the normal distribution of F-actin when zygotes were treated with CH. We will infer from this study and studies in other plant systems that the primary effect of CH was on the disruption of actin-associated cytoskeletal elements. When isolated walls from 8 h zygotes were

* Cytochalasin B, cytochalasin D and dihydrocytochalasin B all have the same general effects and will be referred to collectively in this paper as CH.

Requirements for cell polarity 135

136 R. S. Quatrano and othersstained for sulphated polysaccharides, no localization was observed. By 16 h, how­ever, a clearly delineated segment of rhizoid wall stained positively for sulphated polysaccharides (Novotny & Forman, 1975). Enzymic sulphation of F2 was detected biochemically starting at 10 h. However, embryos continually incubated in CH (no rhizoid formation but some cell division) possessed cell walls that did not contain a stained segment, but showed an evenly distributed stain throughout the wall! All treated zygotes also exhibited an abnormally large accumulation of Golgi and associated vesicles in the perinuclear area. These areas were stained metachromatic- ally with TBO (indicative of sulphated polysaccharides), were capable of incorporat­ing 35S 0 4 into polysaccharides, and persisted as long as CH was present after cell division. Brawley & Quatrano (1979) also observed the wall weakly but uniformly stained with TBO , showing no signs of localized segments of sulphated polysacch­arides. More recently, we demonstrated by particle electrophoresis that the rate of migration of Golgi vesicles from zygotes treated with CH was identical to that of vesicles from untreated embryos (discussed further below). Hence, CH neither inter­fered with sulphation, some secretion of sulphated fucan into the cell wall, nor the surface charge of the Golgi vesicles, but apparently uncoupled these processes from the localized deposition. We have not analysed the walls or cytoplasm from CH- treated zygotes biochemically in order to determine which of the fucan polymers were being sulphated and modified, but assume from cytochemical data that F2 was sulp­hated in a normal manner. These data suggested that sulphation and possibly modif­ication of F 3 to yield sulphated F2 alone did not ensure localization of a fucan polymer. An additional CH-sensitive component (cytoskeleton?) was required.

Since rhizoids do not form in the presence of CH, it must be determined whether the endogenous electrical current was disrupted in CH-treated cells.. In a recent study by Brawley & Robinson (1985), they showed that CH significantly re duced the inward cur­rent at the rhizoid pole (as measured with a vibrating probe). This inhibition was concentration-dependent. CH eliminated the pulsative component of the current and prevented the development of the inward current at the presumptive rhizoid pole. Coupled with their observation that CH caused an altered distribution of F-actin, they suggested that the basis of the CH effect was the disruption of a cyi:oskeletal component.

Surface charge on localized vesiclesWe recently isolated a membrane fraction that contained labelled polysaccharides

after short pulses with 35S 0 4 , had the typical Golgi-marker enzyme inosine diphosphatase (IDPase), was active in secretion, but whose density was considerably higher than typical Golgi preparation from plants. This increase in density was due to an undefined factor in the homogenate. These characteristics i dentified this fraction as the one containing vesicles that became localized in the rhizoid. In addition to these Golgi membranes, an even higher-density membrane fraction accumulated 35S 0 4, did not possess IDPase activity, contained high levels of uronic acid and accumulated rather than secreted 35S0 4 . The developmental and physiological role of this latter fraction remains unclear.

Requirements for cell polarity 137To characterize the surface charge of secretory membranes, two approaches were

used: the electrophoretic mobility (M ) of the vesicles was measured at pH 7, and the isoelectric point (pi) of the vesicles was determined using isoelectric focusing (IEF) (Griffing & Quatrano, 1984). Although both techniques allow derivation of a surface charge value, they gave different information about the ionogenic groups on the membranes since they were done under different conditions. Measurements of M were related to the electrical potential at the plane of shear of the liquid associated with the surface of the moving membrane, whereas IE F measured the potential at the surface directly.

Fig. 3 shows a profile of secretory membranes, which were electrophoresed for 20 min. The membranes from 18 h embryos (Fig. 3 b ) moved faster than those of 20 h Met-embryos (Fig. 3c) showing that, under these conditions, polarly secreted vesicles had a higher negative potential. Interestingly, the 4h embryos had secretory mem­branes that moved at an intermediate rate (Fig. 3 a ) , overlapping extensively with a shoulder of similarly moving vesicles from 18 h embryos. Vesicles isolated from CH- treated embryos had a migration rate identical to those from 18 h embryos (data not shown).

Using IE F , we determined that the pi of the peak of 35S 0 4 -labelled membranes from 20 h Met-embryos was 4-45, significantly higher than the pi from 35S 0 4 -labelled membranes from 18 h embryos, which was measured as 4-1 (Griffing & Quatrano,1984). Again, this showed that secretory vesicles of normal embryos had a higher negative potential than vesicles from Met-embryos. However, the 4 h embryos did not have an intermediate pi under the conditions used for IE F but rather a pi of 4-1, identical to that of normal embryos (data not shown).

The data in Fig. 3 indicated that the Golgi from 4 h and 18 h embryos had a more negative surface potential than Golgi from 20 h Met-embryos. At 4h the zygotes began sulphation and secretion of F 3 and by 18 h F2 was detected. On the other hand, in Met-embryos the fucans were not significantly sulphated (Crayton et al. 1974). Hence, it appeared that sulphation of the fucans was correlated with surface charge. Furthermore, 18 h embryos contained a fucan that had a higher degree of sulphation than 4 h embryos (Hogsett & Quatrano, 1975; Quatrano & Stevens, 1976). This was correlated with a higher negative surface charge in electrophoretic experiments (Fig. 3 a , b ) . It appeared, however, that once the Golgi acquired the capacity to sul­phate at 4 h, the surface charge as assayed by IE F changed very little during develop­ment. This was not surprising since the ionogenic groups detected by IE F were those characteristic of Golgi membranes in general (Griffing & Quatrano, 1984).

Although surface charge was similar on vesicles transported directionally (F 2 at 18 h) or randomly (F3 at 4 h), the absence of F 3 localization could be due either to the lack of a stable endogenous electric field, which is present after 10 h, or to the inability of F 3 vesicles to bind to cytoskeletal or microfilament ‘tracks’. We will use the Nitella system as an assay to determine the ability of Golgi vesicles to bind and to be transpor­ted along actin cables (cf. Sheetz & Spudich, 1983).

138

Distance from origin (m m)

F ig . 3

Requirements for cell polarity 139

SUM M ARY M O D EL FO R L O C A L IZ A T IO N

In 1979 we proposed a working model that was consistent with the available data at that time (Quatrano et al. 1979). Five main points were discussed.

(1) Membrane components responsible for inward current, which leads to a local increase in calcium, were translocated to the presumptive site of rhizoid formation, as determined by an extracellular gradient. Movement of these rhizoid-specific ion channels (RIC S) was mediated by a CH-sensitive event and represented a reversible orientation of the polar axis.

(2) Local increase in ions (such as calcium) at this site may serve to activate enzymes and cellular processes, such as vesicle secretion, microtubule assembly, etc.

(3) R IC S were stabilized at this site by a cytoskeletal component, such as microfila­ments, and represent fixation of the polar axis.

(4) The cytoskeleton may then serve as a track upon which vesicles containing cell wall polymers are transported to the elongating rhizoid, and as a focal point at which to organize and orient the spindle apparatus. Such processes may be modified further by the local increase in specific ions

(5) The directed transport of vesicles could be driven either by an electrical poten­tial gradient (i.e. an electrophoretic mechanism), by a contractile mechanism (i.e. actin-based cytoskeleton), or by both.

The recent data presented by Brawley & Robinson (1985) expand and modify the model for polar axis determination. The localization of cortical F-actin suggested that RICS were normally immobilized and, in the presence of unilateral light, accumulated on the shaded side giving rise to the inward current. The studies of Blatt, Weisenseel & Haupt (1981) on chloroplast aggregation in Vaucheria suggested a similar mechanism. Brawley & Robinson also stressed an amplification component in which the local increase in calcium stimulates F-actin localization and Golgi mem­brane insertions (via secretion) into the rhizoid plasma membrane. Golgi membranes were postulated to contain more R IC S. Finally, they speculated that the cytoskeleton network could trigger local metabolic activation of processes that distinguish between the rhizoid and thallus (e.g. translation of localized rhizoid-specific mRNA), similar to the observations by Moon et al. (1983) in animal systems.

In addition to strengthening the central role of a cytoskeletal component in the establishment of a polar axis, the results presented in this paper point to several other factors that should be included in any model. Post-synthetic modification(s)/ assembly of the sulphated fucan F2 was a critical factor in its polar transport and incorporation into the rhizoid wall only. The modification clearly involved sulphation and possibly the linking, through G-blocks, of ALC side-chains to a fucan sulphate

Fig. 3. Distribution of Golgi vesicles, prepared according to Griffing & Quatrano (1984), containing [35S]fucan after electrophoresis for 20min in a sucrose step-gradient (0-6 to 1*2m) containing 5mM-NaCl in 10mM-barbital buffer (pH 7-2). The vesicles were negatively charged and move from the origin (left) toward the positive pole (right), a . 4 h zygotes; b , 18h embroyos; c, M et-embryos.

140 R. S. Quatrano and othersbackbone. Also, the surface charge on Golgi vesicles was correlated with the levels of sulphation of the enclosed polysaccharide and secretory ability, and was consistent with a self-electrophoretic mechanism of localization. The relationships between the sulphation, assembly and packaging of F2 into Golgi vesicles, the surface properties of these vesicles (other than charge) and their association with a cytoskeletal com­ponent must now be elucidated to gain a more complete understanding of how F2 is polarly secreted.

Research from our laboratory was supported by funds from The National Science Foundation (PCM 78-20435). This report was prepared while R. S. Q. was on a NSF U.S./U.K. Cooperative Fellowship at the Plant Breeding Institute, Cambridge, England.

R E F E R E N C E S

B l a t t , M. R., W e ise n se e l, M. H. & H aupt, W. (1981). A light-dependent current associated with chloroplast aggregation in the alga Vaucheria sessilis. Planta 152, 513-526.

B raw ley , S.H. & Q u a tra n o , R. S. (1979). Sulfation of fucoidin in Fucus embryos. IV. Autoradiographic investigations of fucoidin sulphation and secretion duri ng differentiation and the effect of cytochalasin treatment. Devi Biol. 73 , 193-205.

B ra w ley , S. H. & R obinson , K. R . (1985). Cytochalasin treatment disrupts the endogenous currents associated with cell polarization in fucoid zygotes. Studies of the role of F-actin in embryogenesis. J . Cell Biol, (in press).

C ra y to n , M. A., W ilso n , E. & Q u a tra n o , R. S. (1974). Sulphation of fucoidan in Fucus embryos. II . Separation from initiation of polar growth. Devi Biol. 39 , 134-137.

D o u b e t, R. S. (1983). Synthesis and assembly of Fucus zygote cell walls with an emphasis on alginate and fucans. Ph.D. dissertation, Oregon State University.

D o u b e t, R. S. & Q u a tra n o , R. S. (1984). Properties of alginate lyases from marine bacteria. Appl. environ. Microbiol. 47 , 699—703.

Evans, L. V., C a llo w , J . A. & C a llo w , M. E . (1982). The biology and biochemistry of reproduction and early development in Fucus. Prog. Phycol. Res. 1, 67—110.

G ra sd a le n , H., L a rse n , B. & Sm idsrod, O. (1981). Studies on the sequence of uronic acid residues in alginic acid. Acta chem. scad. 89, 179-191.

G r if f in g , L . & Q u a tra n o , R. S. (1984). Isoelectric focusing of plant cell membranes. Proc. natn. Acad. Sci. U SA . 81 , 4804-4808.

H o g s e tt , W. E. & Q u a tra n o , R. S. (1975). Isolation of polysaccharides sulphated during early embryogenesis in Fucus. PI. Physiol. 55, 25-29.

H o g s e tt , W. E. & Q u a tra n o , R. S. (1978). Sulphation of fucoidin in Fucus embryos. III. Required for localization in the rhizoid wall. J . Cell Biol. 78, 866-873.

J a f f e , L. F. (1968). Localization in the developing Fucus egg and the general role of localizing currents. Adv. Morphogen. 7 , 295-328.

J a f f e , L. F. & N u c c i t e l l i , R. (1977). Electrical controls of development. A. Rev. Biophys. Bioengng. 61 , 445-476,

M ian, A .J. & P e rc iv a l, E. (1973). Carbohydrates of the brown seaweeds Himanthalia lorea and Bifurcaria bifurcata. Part II. Structural Studies of the ‘Fucans’. Carbohyd. Res. 26 , 147-161.

M e d c a lf , D. G. & L a rse n , B. (1977a). Fucose-containing polysaccharides in the brown algae Ascophyllum nodosum and Fucus vesiculosis. Carbohydr. Res. 59 , 531-527.

M e d c a lf , D. G. & L a rse n , B. (19776). Structural studies on ascophyllan and the fucose- containing complexes from the brown algae Ascophyllum nodosum. Carbohydr. Res. 59 , 538-545.

M oon, R. T . N ico sia , R. F ., O lse n , C., H il le , M . B. & J e f f r e y , W. R. (1983). The cytoskeletal framework of sea urchin eggs and embryos: developmental changes in the association of messen­ger RNA. Devi Biol. 95, 447-458.

N ovotn y , A. M. & Form an, M. (1975). The composition and development of cell walls of Fucus embryos. Planta 122, 67—78.

Requirements for cell polarity 141N u c c i t e l l i , R. (1978). Ooplasmic segregation and secretion in thePelvetia egg is accompanied by

a membrane-generated electrical current. Devi Biol. 62, 13-33 .Q u a tra n o , R. S. (1973). Separation of processes associated with cellular differentiation in two-

celled Fucus embryos. Devi Biol. 30 , 209-213.Q u a tra n o , R. S. (1978). Development of cell polarity. A. Rev. PI. Physiol. 29 , 487-510.Q u a tra n o , R. S. (1980). Gamete release, fertilization and embryogenesis in the Fucales. In

Handbook o f Phycological Methods, Vol. I l l , Developmental and Cytological Methods (ed. E. Gantt), chap. 6, pp. 59—68. Cambridge University Press.

Q u a tra n o , R. S., B raw ley , S. H . & H o g s e tt , W. E. (1979). The control of the polar deposition of asulphated polysaccharide in Fucus zygotes. In Determinants o f Spatial Organization (ed. S. Subtelny & I. R. Königsberg), pp. 77-96. New York: Academic Press.

Q u a tra n o , R. S. & C ra y to n , M. A. (1973). Sulphation of fucoidan in Fucus embryos. II. Possible role in localization. Devi Biol. 30 , 29-41.

Q u a tra n o , R. S . & S te v e n s , P. T . (1976). Cell wall assembly in Fucus zygotes. I. Character­ization of the polysaccharide components. PI. Physiol. 58, 224—231.

R ob inson , K. R . & J a f f e , L. F. (1975). Polarizing fucoid eggs drive a calcium current through themselves. Science 187, 70-72.

S h e e tz , M. P. & Spudich, J. A. (1983). Movement of myosin-coated fluorescent beads on actin cables in vitro. Nature Lond. 303 , 31-35.

V r e u g d e n h il l , D ., D i jk s tr a , M. L . & L ib b en g a, K. R. (1976). The ultrastructure of the cell wall of normal and polar embryos of Fucus vesiculosus. Protoplasma 88 , 305-313.