Plant Cell Physiol. 48(1): 122–133 (2007) doi:10.1093/pcp/pcl053, available online at www.pcp.oxfordjournals.org ß The Author 2006. Published by Oxford University Press on behalf of Japanese Society of Plant Physiologists. All rights reserved. For permissions, please email: [email protected] Spectroscopic and Biochemical Analysis of Regions of the Cell Wall of the Unicellular ‘Mannan Weed’, Acetabularia acetabulum
Erin K. Dunn 1, 5, Douglas A. Shoue 2, 5, Xuemei Huang 3, Raymond E. Kline 4, Alex L. MacKay 3, Nicholas C. Carpita 2, Iain E.P. Taylor 4 and Dina F. Mandoli 1, 1 Department of Biology, Center for Developmental Biology & Institute for Stem Cell and Regenerative Medicine, Box 35325, University of Washington, Seattle, WA 98195-5325, USA 2 Department of Botany and Plant Pathology, Purdue University, W. Lafayette, Indiana 47907-2054, USA 3 Department of Physics and Astronomy, University of British Columbia, Vancouver, BC, Canada V6T 1Z1
4 Department of Botany, University of British Columbia, Vancouver, BC, Canada V6T 1Z4 Downloaded from https://academic.oup.com/pcp/article/48/1/122/2469303 by guest on 29 September 2021
Although the Dasycladalean alga Acetabularia acetabulum Introduction has long been known to contain mannan-rich walls, it is not known to what extent wall composition varies as a function of The Mediterranean coralline alga Acetabularia the elaborate cellular differentiation of this cell, nor has it acetabulum (previously classified as A. mediterranea)isa been determined what other polysaccharides accompany the giant unicell with a precisely defined, intricate morphology. mannans. Cell walls were prepared from rhizoids, stalks, This diploid organism produces an upright stalk from which hairs, hair scars, apical septa, gametophores and gametangia, whorls of branched hairs emerge, a basal rhizoid composed subjected to nuclear magnetic resonance and Fourier trans- of multiple digits which houses the cell’s sole nucleus for form infrared spectroscopy, and analyzed for monosaccharide most of the life cycle, and later in development an elaborate composition and linkage, although material limitations reproductive structure or ‘cap’ at the apex, 3–4 cm away prevented some cell regions from being analyzed by some of from the rhizoid (Fig. 1). During reproduction, haploid the methods. In diplophase, walls contain a para-crystalline nuclei are generated, which migrate with the majority of the mannan, with other polysaccharides accounting for 10–20% parental cytoplasm into the cap. Once there, each nucleus of the wall mass; in haplophase, gametangia have a cellulosic along with some cytoplasm is surrounded by a new cell wall, wall, with mannans and other polymers representing about a creating the ‘gametangia’. The subsequently produced quarter of the mass. In the walls of the diplophase, the gametes lack a cell wall of any kind. Previous studies have mannan appears less crystalline than typical of cellulose. indicated that the vegetative wall of this organism consists The walls of both diploid and haploid phases contain little mostly of mannan, but that during reproduction the if any xyloglucan or pectic polysaccharides, but appear to gametangial wall is about 80% cellulose and about 20% contain small amounts of a homorhamnan, galactomannans mannan (Zetsche et al. 1970). and glucogalactomannans, and branched xylans. These There are three reasons why further investigation of the ancillary polysaccharides are approximately as abundant wall of A. acetabulum is of interest. First, the cell comprises in the cellulose-rich gametangia as in the mannan-rich more than stalk, cap and gametangia—prior studies ignored diplophase. In the diplophase, different regions of the cell the other distinct regions of the cell (e.g. rhizoid, hairs, differ modestly but reproducibly in the composition of the cell septa; Fig. 1). Indeed, the ‘stalk’ and ‘cap’ in previous wall. These results suggest unique cell wall architecture for studies are themselves heterogeneous (Fig. 1). It is reason- the mannan-rich cell walls of the Dasycladales. able that the morphological and physiological differentia- tion of the cell may be accompanied by differences in cell Keywords: Acetabularia acetabulum — Cellulose — wall composition and structure. However, such spatial Dasycladales — FTIR — Linkage analysis —Mannan differences in cell wall composition have been little studied weed. in any giant unicellular species. Here, we carefully dissected distinct regions of the cell for analysis. Abbreviations: FTIR, Fourier transform infrared spectro- Secondly, mannans in this type of cell wall have been scopy; GC–MS, gas chromatography–mass spectrometry; GalA, galacturonic acid; GlcA, glucuronic acid; ManA, mannuronic acid; poorly characterized compared with cellulose, leaving NMR, nuclear magnetic resonance; RG I, rhamnogalacturonan I; doubts about the form(s) of mannan present. In theory, TFA, trifluoracetic acid. a microfibril made from mannans would be all but identical
5Present address: c/o University of California San Diego, Biology Graduate Student Affairs, 0348, 9500 Gilman Drive, La Jolla, CA 92093-0348, USA. Corresponding author: E-mail, [email protected]; Fax, þ1–206–685–1728.
122 Cell walls of Acetabularia 123
Gametophores Cap Gametangial wall encloses Cap septum each nucleus Mitoses (~9x)
Whorl scar Downloaded from https://academic.oup.com/pcp/article/48/1/122/2469303 by guest on 29 September 2021 Hairs Stalk Interwhorl
Gametophore
Nucleus Rhizoid (2n) Meiosis (1x) then mitoses (~9x) followed by transport of haploid Gametangium nuclei into the cap
Fig. 1 Anatomy of vegetative adult and reproductive individuals. Septa join the stalk to the cap and each hair to the stalk. Older hairs are normally shed during vegetative growth. When hairs separate from the stalk, they leave behind a ‘whorl scar’. Cap initiation marks reproductive onset. Later, during cap expansion, meiosis of the single diploid nucleus and subsequent rounds of mitosis generate thousands of haploid nuclei. For clarity, not all the nuclei, gametangia and gametes were drawn. Hairs are usually completely shed by the time that the cap bears gametangia. To clarify anatomy, images are drawn only roughly to scale. Cells average 3–4 cm tall, caps average 1 cm and stalks average 0.3 mm in diameter. Detailed dimensions of anatomical regions within cell populations can be found in Nishimura et al. (1992a, b) and also in Ngo et al. (2005).
to cellulose; however, a cellulose-like high tensile strength is depends on the interactions between cellulose and pectin generally not ascribed to the mannan-rich walls of (Proseus and Boyer 2006), with xyloglucan being absent A. acetabulum or of other species with mannan-rich walls (Popper and Fry 2003). How the major mannan phase is (the so-called ‘mannan weeds’). Were the mannan sub- modified by the remaining cell wall matrix has been almost stituted to even a small extent, then this would undermine totally unexplored. its ability to form high tensile strength microfibrils, but Here, we have used Fourier transform infrared might enhance its ability to form a flexible network. spectroscopy (FTIR), nuclear magnetic resonance (NMR), Frei and Preston (1968), using X-ray diffraction, observed and sugar and methylation analyses to characterize each physical polymorphism within the mannan from whole region of the wall of A. acetabulum. We found that the A. acetabulum walls but did not find microfibrillar forms. mannan makes up almost 90% of the diplophase cell wall Chanzy (1984), using electron diffraction and microscopy, mass and that the diplophase contained small amounts of confirmed the existence of at least two forms of mannan hemicelluloses, resembling those described for other cells but after chemical extraction, but detected both granular and not including xyloglucan, and had essentially no recognizable microfibrillar domains. Neither author prepared cell walls pectin. The haplophase, although containing a majority of from defined regions of the cell. cellulose, had remaining cell wall material similar to that of the Thirdly, although the identity of the major cell wall diplophase rather than to the cellulosic walls of other algae. component of diplophase (rhizoid, stalk, cap) and haplo- phase (gametangia) has been identified as mannan and Results cellulose, respectively, there has been almost no analysis of the other polymers of the cell wall. Even algal orders that Regions of the vegetative and reproductive thalli dissected for contain cellulose-rich walls, such as charophytes, have non- analysis cellulosic components that differ from that found in the The anatomy of both vegetative and reproductive thalli type I or type II walls of angiosperms. It has recently of this unicell is complex. Fig. 1 is a labeled drawing of become clear, for example, that elongation in Chara corallina subcellular regions of the wall that are macroscopically 124 Cell walls of Acetabularia
Table 1 Amino acid composition (mean% of totals) in stalk and cap walls of A. acetabulum Amino acid Asx Glx Lysine Leucine Totals of other amino acidsa Stalks 12.3 12.5 13.9 6.4 54.9 Caps 14.1 16.0 8.5 9.0 52.4 Values are expressed as a percentage of the total amino acids recovered in analysis of that region of the thallus wall. a Total other amino acids includes threonine, serine, glycine, alanine proline methionine, valine, isoleucine, tyrosine, phenylalanine, histidine and arginine, which each were always58% of the total amino acids recovered. Asx ¼ (Asn þ Asp) and Glx ¼ (Gln þ Glu). We did not detect tryptophan, hydroxyproline or cysteine. Downloaded from https://academic.oup.com/pcp/article/48/1/122/2469303 by guest on 29 September 2021 distinct and could be dissected unambiguously and drawn 1065
Gametophores 1078 roughly to scale. At no time were haplophase and A. 1038 diplophase walls mixed, i.e. gametangia were always manually removed from gametophores prior to analysis. 1165 In all experiments, we dissected to the smallest possible 1146 1223
region within the limitations of the technique. The regions 1261 1380 used are specified for each technique. Details of the 1624 1414
dissection process and of cell wall preparation are given in 1060 Materials and Methods. 1038
B. Gametangia 1107 Quantitation of amino acids The limited supply of materials (recall that these are 1161 1319 1635 single cells) allowed us to make only semi-quantitative 1373 1427 1616 1338
observations about amino acid and amino sugar distribu- 1254 1543 tion (Table 1). We had sufficient mass to study the two 1736 major wall types made during development: ‘vegetative C. Gametophores regions’ (interwhorls plus whorl scars but no hairs and no 1215 1180
minus Gametangia 1078 rhizoids) and ‘reproductive regions’ (cap septum plus 1142 1011
gametophores). We found that amino acids and amino 1415 1389 sugars constituted approximately 1–2% of the total wall 1616 1417
mass in each of these regions. Consistent with earlier reports 1161 1111 982 1734 903 833 1589
(see Taylor and Cameron 1973) from other species, 1543 1335 1319 1658 we found that asparagine/aspartate, glutamine/glutamate, 1800 1600 1400 1200 1000 800 lysine and leucine were the predominant amino acids. The Wavenumber (cm−1) hydrolysis (6 N HCl) and amino acid analyzer procedures (ninhydrin) were suitable to detect all amino acids, except Fig. 2 FTIR spectra obtained from isolated wall fragments of the asparagine and glutamine, which are deaminated to mannan-rich (A) gametophores and (B) cellulose-rich gametangia. The difference spectrum (C) of gametophores minus gametangia aspartate and glutamate, respectively, and hydroxyproline, was obtained from digital subtraction after area averaging and which required a colorimetric analysis and was not detected baseline correction of the original spectra A and B. Absorbance in this study. Tryptophan and cysteine were also not peaks enriched in the mannan wall are positive (labeled with the detected. wavenumber above the spectrum), and those enriched in the cellulosic wall are negative (labeled with the wavenumber below the spectrum). FTIR microspectroscopy distinguishes mannan- and cellulose-rich cell wall regions FTIR is a non-invasive, non-destructive method that the mannan-rich gametophores and cellulose-rich gametan- provides information about both structure and bonding of gia (Fig. 2). The carbohydrate fingerprint region of the the organic molecules being analyzed (McCann et al. 1997). spectrum of gametangial walls was dominated by absor- With sampling areas as small as a few hundred microns, bance maxima at 1,161, 1,107, 1,060 and 1,038 cm 1, chemical groups can be mapped across the morphological consistent with the wavenumbers associated with cellulose features of a large unicell. FTIR spectra were obtained from (Figure 5 in Tsuboi 1957, Liang and Marchessault 1959). three 150 mm 150 mm regions of A. acetabulum walls from In contrast, the mannan-rich walls of the gametophores Cell walls of Acetabularia 125 absorbed maximally at 1,078 and 1,146 cm 1 in addition to Using NMR, we measured three physical parameters, having peaks typical of cellulose, indicating that the M2,M2interpair and T1D, because each parameter gives gametophore walls contained both mannan and cellulose different kinds of information about wall structure. The (Kacura´kova´et al. 2000). Because of the manner in which interpretation of such broadline NMR quantities in terms the regions of wall were isolated, both preparations of cell wall microstructure has been detailed elsewhere contained protein, which yields amide absorbance maxima (MacKay et al. 1985, MacKay et al. 1988). 1 in a broad range from 1,630 to 1,650 cm and from 1,540 The second moment of the NMR line shape, ‘M2’, is a to 1,550 cm 1. Absorbances typical of carboxylate ester measure of the orientational order of the polymer molecules 1 1 (1,736 cm ) and carboxylate ions (1,616 and 1,437 cm ) in these wall regions. M2 is large for cell wall preparations are more intense in the gametangial wall than in the from other species known to contain highly crystalline Downloaded from https://academic.oup.com/pcp/article/48/1/122/2469303 by guest on 29 September 2021 gametophore wall. cellulose. None of our samples had M2 values that approach A difference spectrum (Fig. 2c), made by digitally the values observed for crystalline cotton cellulose subtracting the area-normalized spectra of two physically (7.3 109 s 2), which we used as a reference for fully purified wall types (Fig. 2a, b), revealed that the key restrained molecular motion (MacKay et al. 1988). features that distinguished gametophore from gametangial Nevertheless, there was considerable orientational order in walls are at wavenumbers specific to mannan-rich vs. all four preparations of A. acetabulum cell walls (Table 2). 9 2 cellulose-rich cell walls. Whereas wavenumbers 1,142 and The M2 values for walls from fresh cap (5.2 10 s ) and 1,078 cm 1 (positive peaks) were enriched in the gameto- stalk (4.2 109 s 2) indicated more molecular motion than phore, wavenumbers 1,161 and 1,111 cm 1 (negative peaks) the cotton cellulose but similar motion to that observed in were enriched in gametangia walls, consistent with assign- the bean cell walls studied earlier (MacKay et al. 1985, ments of mannan-rich and cellulose-rich from the sugar MacKay et al. 1988). Comparison of M2 results also analyses. Enrichment in the gametangia for wavenumbers demonstrated that lyophilization of fresh stalk walls characteristic of uronic acid carboxylate esters and carbox- increased the M2 (more order) but that rehydration reversed ylate ions were also observed. The difference spectrum also this effect. indicated enrichment of absorbance at 1,415, 1,389, 1,215 Signals from of these four wall preparations were and 1,180 cm 1 in mannan-rich walls, and 1,335 and resolved into two physical components based on both 1 1,319 cm in the cellulose-rich gametangial walls, but the M2interpair and T1D measurements. Based on prior work, we absorbances at these frequencies were not well defined. interpreted M2interpair as a measure of molecular motion (or lack thereof) and T1D as a measure of crystallinity (MacKay et al. 1985, MacKay et al. 1988). The solid echo measure-
NMR reveals that both crystalline and non-crystalline ment provided M2interpair values that distinguish wall mannan is present subcomponents on the basis of molecular motions:
To assess the structural disposition of the polymers in a larger M2interpair value corresponds to a more rigid the A. acetabulum cell wall, we used NMR. Crystallinity and structure. In a study of the cellulosic wall of bean, the rigidity were estimated from the second moment, solid echo component with the larger M2interpair was assigned to all the and Jeener echo. We dissected the thallus into vegetative cellulose and some of the hemicellulose molecules (the more and reproductive regions: a length of whole late adult stalks rigid hemicellulose molecules), while the component with
(interwhorl plus whorl scars but no rhizoids or hairs, which the smaller M2interpair was assigned to the less rigid pectin had fallen off earlier in development) and whole cap walls and hemicellulose (MacKay et al. 1988). The wall regions of (cap septum plus gametophores). Gametangia were not A. acetabulum also contained two distinguishable compo- analyzed. These portions of the thallus were either nents, with M2interpair values differing by about one order of (i) lyophilized only; (ii) lyophilized and then rehydrated; magnitude. Drying before measurement altered the relative or (iii) dried manually by squeezing between glass fiber filter proportions of these two M2interpair components, apparently paper, and (iv) air-dried (Table 2). Each of these were stiffening some of the wall solids (Table 2). Unlike distinct preparations rather than sequential treatments of the cellulosic walls of bean (MacKay et al. 1982), in the same regions. Comparison of these types of wall A. acetabulum this stiffening was completely reversed preparation was motivated by earlier work in which it was by rehydration (Table 2). discovered that drying bean (Phaseolus vulgaris) cell walls For A. acetabulum, as in bean (MacKay et al. 1988), resulted in a flexible component that became rigid and the Jeener echo measurement revealed two components with remained so after rehydration (MacKay et al. 1982). different T1D values. Following the previous assignment for We wanted to learn if wall regions of A. acetabulum had longer T1D, 11–13% of the protons in the wall regions of a similar component and were therefore subject to the same A. acetabulum were from crystalline polysaccharide. experimental artifact(s). In contrast, in bean, about 70% of the cellulose was 126 Cell walls of Acetabularia
Table 2 NMR analysis of cell walls from A. acetabulum NMR parameter Vegetative Reproductive structure, structures, all gametophores only Lyophilized Lyophilized and Fresh walls Fresh walls and ground ground to powder, pressed between air-dried to powder then rehydrated glass fiber
with D2O filter paper 9 2 M2,10 s 5.1 4.2 4.2 5.2 a
Solid echo (M2interpair) , Downloaded from https://academic.oup.com/pcp/article/48/1/122/2469303 by guest on 29 September 2021 109 s 2 Component A 4.7 (96%) 3.9 (70%) 5.5 (61%) 4.5 (74%) Component B 0.7 (4%) 0.4 (30%) 0.6 (39%) 0.18 (26%) b Jeener echo (T1D), ms Component A 20 (13%) 30 (11%) 10 (11%) Decayed quickly, both signals gone at 5 ms Component B 2.5 (87%) 10 (89%) 2 (89%) For each method of wall preparation, the values in parentheses calculate the percentage of the wall polysaccharides accounted for by that parameter. These percentages total to 100%. a M2interpair is a measure of the decay of the solid echo. A smaller M2interpair indicates more molecular motion. In all cases, two M2interpair components were measured. b 5 T1D, the dipolar relaxation time, is a measure of vibrations, which occur in the order of 10 s and is measured from the decay of the Jeener echo. The component with the longer T1D is assigned to crystalline wall. In all vegetative structures, two T1D components were measured.
crystalline (MacKay et al. 1988). The proportion of this About three-quarters of the mass of the cellulosic game- crystalline component was not altered by the method of tangial cell walls was acid resistant, and for the mannan-rich drying (Table 2), but the results from the caps were not walls of the diplophase the percentage of acid-resistant interpretable because the T1Ds were far shorter than for material exceeded 80% (Table 3). The insoluble material the other regions, i.e. their decay was very fast (Table 2). was not analyzed further. These observations suggest that these reproductive regions In TFA-soluble fractions, mannose varied from 82 to contained minimal crystalline polysaccharide. 95 mole% of the hydrolyzable polysaccharide in the The M2interpair results indicate that between 61 and diplophase (i.e. gametophores, cap septa, interwhorls, 74% of the hydrated walls consisted of molecules that have hairs, whorl scars and rhizoids), whereas mannose was some molecular rigidity, while the remaining material only about 32 mole% of soluble material from gametangial contained molecules that undergo much more motion. By walls (Table 3). In both diplophase and gametangia, glucose analogy to previous analysis of cellulose in the cell walls of accounted for most of the remainder, and four other bean (Taylor et al. 1990), we assigned the component with monosaccharides, namely rhamnose, arabinose, xylose and the longer T1D, which amounted to 12% of the total, to galactose, were present in smaller amounts. Glucose the crystalline polysaccharide in A. acetabulum. Given that probably comes from amorphous cellulose as the amounts chemical analysis showed that the stalk and cap are at least of glucose recovered in TFA hydrolysis are corrected 80% mannan and a signal in NMR requires that a given with the amounts of cellulose detected. From carboxyl polysaccharide be about 20–30% of the total mass, we reduction with sodium borodeuteride detected by gas suggest that most (61–74%) of the mannan is relatively chromatography–mass spectrometry (GC–MS), we found rigid, but only about 10% of it is actually crystalline. trace amounts of the sugar acids, galacturonic acid (GalA) and glucuronic acid (GlcA), but mannuronic acid (ManA) Quantification of monosaccharides and polysaccharide was undetectable. linkages The proportion of galactose, rhamnose, arabinose and To assess the composition of the A. acetabulum cell xylose differed somewhat among the walls from different wall, we assayed monosaccharide abundance and linkages. cell regions (Table 4). These sugars had about the same Cell walls were purified, hydrolyzed in 2 M trifluoroacetic abundance in whorl scars and rhizoids, with galactose being acid (TFA), and soluble and insoluble fractions collected. the most abundant. Gametophores and the neighboring Cell walls of Acetabularia 127
Table 3 Proportions of acid-resistant material in the three principal body regions, and neutral sugar distribution in the acid-soluble fraction of seven differentiated regions of the wall Gametophores Interwhorls Cap septa Hairs Rhizoids Whorl scars Gametangia Mean mole% of glucose or mannose equivalents 2 M TFA-resistant sugar 81.5 0.28 6.1 3.9 [–] [–] [–] [–] 73.5 1.4 Mean mole% 2 M TFA-sensitive sugar Rhamnose 1.8 0.3 3.4 4.3 0.6 0.1 2.0 Arabinose 0.2 0.3 0.4 0.8 0.3 0.2 0.5 Downloaded from https://academic.oup.com/pcp/article/48/1/122/2469303 by guest on 29 September 2021 Xylose 0.6 0.4 0.4 1.5 0.3 0.2 6.4 Mannose 86.1 91.4 83.3 82.3 92.4 94.5 32.3 Galactose 4.3 3.6 8.5 0.7 4.4 3.7 3.3 Glucose 6.2 3.8 4.0 10.3 2.1 1.3 55.3 GalA tr tr 0.1 [–] [–] [–] 0.1 GlcA 0.8 [–] [–] [–] [–] 0.1 Values are the mean of at least two independent samples, with a maximum variance of 510%. tr, trace amounts 50.05%; [–], not determined. Values for GlcA and GalA in the caps, stalk and gametangia were determined from proportions of 6,6-dideutero alditol acetate derivatives in their respective neutral sugar by GC–MS. Amounts were calculated as described by Kim and Carpita (1992) after correction for 13C spillover.
Table 4 Linkage composition (mole%) of cell wall polysaccharides from various cell regions Linkage Interwhorls Gametophores Rhizoids Cap septa Gametangia 2-Rhap 0.3 1.8 0.6 3.1 2.0 5-Araf 0.3 0.2 tr tr 0.5 4-Xylp 0.3 0.4 0.2 0.4 6.0 2,4-Xylp 0.1 0.2 0.1 0.4 0.3 3,4-Xylp tr tr tr 0.2 0.1 t-Galp 3.6 4.3 4.4 8.9 3.3 t-Manp 5.1 4.9 2.4 0.6 0.3 2-Manp tr 2.0 4.3 1.0 tr 4-Manp 77.9 69.0 60.3 72.1 25.5 6-Manp 2.1 0.6 10.2 tr tr 2,4-Manp 2.0 3.1 4.4 3.3 2.3 3,4-Manp 1.5 2.8 5.5 2.5 1.5 4,6-Manp 2.9 3.7 5.3 tr 2.7 t-Glcp 0.3 0.1 0.1 tr 12.4 2-Glcp n.d. tr 0.1 0.5 n.d. 3-Glcp n.d. n.d. n.d. n.d. 2.0 4-Glcp 3.2 5.7 1.5 6.4 30.2 6-Glcp 0.3 0.1 0.1 n.d. 2.5 2,4-Glcp tr tr 0.1 tr 1.5 3,4-Glcp tr 0.1 n.d. tr 3.6 4,6-Glcp tr 0.2 0.3 0.6 3.1 t-GlcAp 0.1 0.8 [–] [–] 0.1 t-GalAp tr tr [–] [–] tr 4-GalAp tr tr [–] [–] 0.1 tr, trace amounts between 0.01and 0.05 mole%; n.d., not detected; [–] ¼ not determined. Linkage is deduced by the yield of fragments upon electron impact mass spectrometry of partially methylated alditol acetate derivatives separated by gas–liquid chromatography. Thus, 4-Glcp indicates the attachment of another residue at the O-4 position of glucopyranose. The actual derivative is 1, 4, 5-triacetyl(1-deutero)-2,3,6-methyl-glucitol. 128 Cell walls of Acetabularia
septa had similar abundances, only here rhamnose technique. The substantial amounts of non-reducing term- increased in abundance though not exceeding galactose. inal mannose found in some regions could account for However, in hairs, rhamnose was the most abundant and branch termini, but the t-Galp could also contribute galactose was the least abundant, a sequence that differs termini. Despite the dominance of mannan, significant from the other regions. In gametangia, despite the replace- amounts of 4-linked glucose were detected but no other ment of mannan with cellulose as the major cell wall linkages, indicating that cellulose may co-exist with the polysaccharide, the distribution of the minor sugars was mannan. The trace amounts of 4,6-Glcp and lack of t-Xylp similar to that of the gametophores, with approximately residues indicates that mannan-rich walls of A. acetabulum similar amounts of galactose, rhamnose and xylose, and less lack xyloglucan, the major cellulose cross-linking glycan of arabinose. vascular plant cell walls. Downloaded from https://academic.oup.com/pcp/article/48/1/122/2469303 by guest on 29 September 2021 To assess the kinds of polysaccarides present in the In contrast to the mannan-rich wall regions of the A. acetabulum cell wall, we analyzed the linkages present in diplophase, the gametangial wall was dominated by the the monosaccharides released from the wall by TFA 4-Glcp residues typical of cellulose. Substantial amounts of hydrolysis. Linkage analysis showed that all wall regions, all three possible 4-Glcp branch point residues occurred, except the interwhorl and rhizoid, had significant amounts but mostly as 3,4- and 4,6-linked residues. As with the of 2-Rhap but no detectable 2,4-Rhap branch point residues mannan wall regions, t-Xylp residues were absent, suggest- (Table 4). Although in trace amounts in the gametophore ing that xyloglucan typical of vascular plants was lacking. and interwhorl regions, and barely detectable in the The candidate terminal residues corresponding to these gametangial wall, both terminal- and 4-linked GalAp branch points are glucose and/or galactose. There is residues were detected by mass spectrometry of the insufficient t-Man, t-GlcA or t-GalA to account for the diagnostic, partly methylated alditol acetate derivatives. degree of branching observed. Xylans contributed more GlcA is present in small amounts and found solely as non- substantially to the gametangial wall than to that of the reducing terminal residues. The small amounts of arabinose diplophase, but they were less branched. Similarly, 2-Rhap were 5-linked in all wall regions: although 5-Araf gives the and 4-GalAp rose to their highest proportions in the same MS fragmentation as 4-Arap, these are resolved by gametangial wall. Mannan was a significant wall poly- elution time in the GC (Carpita and Shea 1989). Xylose in saccharide in the gametangial wall, with mannose all regions appeared to be from a 4-linked xylan, branched contributing about 32 mole% of the residues. Among at primarily the O-2 position, although only gametangial mannose residues, the three possible branch points were walls contained substantial amounts of this polysaccharide. detected in roughly the same proportions as in the mannan- Based on the ratio of 4-Xylp : 2,4-Xylp, the xylans were rich walls. highly branched in the cap septum, moderately so in the Total uronic acids, assayed colorimetrically, consti- gametophore and rhizoid, and much less so in gametangia. tuted 51% of the monosaccharide of the gametophore and Although it is theoretically possible that mannan has its stalk regions, and of the gametangial walls. When the backbone interrupted by xylose residues, there is no uronic acid residues were carboxyl reduced with a water- precedence for this. In plants, glucomannan is the only soluble diimide and NaBD4, GlcA was the predominant known polysaccharide with a mixed 4-linked backbone, uronic acid and was generally accompanied by trace typically comprising short chains of each, not single amounts of GalA (Table 2). GalA is generally 10-fold residues. Further argument against an association with or less abundant than rhamnose. No ManA was detected mannan is the fact that 4-Xyl is most abundant in the in any of these major wall regions. gametangial wall. The galactose residues in all wall regions were non-reducing terminal units. Non-reducing terminal Discussion deoxysugars and pentoses were not found in any wall region, but both t-Manp and t-Glcp were detected in Although the cell wall composition of many algal all samples, with higher percentages in their respective groups is uncharacterized, mannan is the predominant glycan-enriched walls. skeletal wall polymer in several genera of green algae, For mannan-rich walls, the majority of mannose was including Acetabularia (Iriki and Miwa 1960, Chanzy et al. 4-linked. Other linkages were present in small amounts, in 1984) in the Dasycladales and Codium in the Caulerpales proportions unique to the gametophore, rhizoid or cap (Graham and Wilcox 2000). This is sufficiently unusual that septum. The three possible branched forms of 4-linked algae in these genera are often referred to as ‘mannan mannose, namely the 2,4-, the 3,4- and the 4,6-Manp weeds’. In both genera, the mannan wall is made during the residues, were found in amounts that varied from trace to vegetative phase but a predominantly cellulosic wall is made 45 mole% of the total sugar, indicating that they were not during the reproductive phase. To the best of our knowl- products of undermethylation, i.e. attributable to poor edge, nothing is known about how such a major shift in wall Cell walls of Acetabularia 129 biosynthesis is accomplished. In addition, while previous crystalline mannan was looser than the packing of crystal- work documented mannan as the predominant polymer in line cellulose. The M2 of the dried A. acetabulum wall the vegetative phase of these taxa, the remaining compo- powder was similar to (but slightly less than) the value nents of the wall have been little analyzed, so it is not obtained from dried bean cell walls (MacKay et al. 1988), known which, if any, components are the counterparts to but lower than that of crystalline cellulose from cotton the pectin and hemicellulose of higher plant cell walls. (MacKay et al. 1985). The M2 values of the rehydrated and In addition, this is apparently the first time that the cell wall the fresh, untreated A. acetabulum walls were less than composition of distinct, differentiated regions of a uni- that of dried walls, suggesting that water loosened the cell cellular algal cell wall have been analyzed. By combining wall polymers. Previous NMR studies of bean cell walls
FTIR microspectroscopy, NMR and biochemistry, we assigned the component with the larger M2interpair to demonstrate that cell walls of the diploid thallus of motionally restricted molecules, such as cellulose and Downloaded from https://academic.oup.com/pcp/article/48/1/122/2469303 by guest on 29 September 2021
A. acetabulum contain novel neutral polysaccharides but ordered associated polymers, and the smaller M2interpair little pectin and that the wall chemistry within varies with component to the less motionally restricted molecules, local morphology. These walls also contain 1% protein, namely hemicellulose and pectin (MacKay et al. 1988). far less than that in cell walls of vascular plants. The slight Taken together, these data suggest that this mannan wall variation in amino acid composition of isolated walls of is more flexible than a cellulosic one. vegetative and reproductive cells hints that distinct regions The amount of acid-resistant material was correlated of the wall contain different amounts of certain structural with mannose abundance; the mannan-rich interwhorl wall proteins. Taken together, our data suggest that the structure was almost 90% resistant to TFA, and soluble mannose of the mannan-rich cell wall involves more than a constituted up to 95% of the monosaccharide, whereas straightforward substitution of mannan for glucan micro- gametophore wall had less resistant crystalline mannan and fibrils and that these giant unicells contain spatial as well as less mannose. Likewise, gametangial walls were least temporal differentiation in cell wall biosynthesis. resistant to acid hydrolysis and had mixtures of predomi- nantly glucose and mannose, indicating a possible inter- Apparent crystallinity of the mannan weaving of polymers with less crystallinity. Para-crystalline refers to a crystalline core resistant to strong acid (such as TFA) surrounded by less organized Polysaccharides vary with thallus architecture chains that exhibit some sensitivity to hydrolysis in the TFA In A. acetabulum, the diplophase wall (i.e. gameto- (seen as mannose and glucose in their relative proportions phore, cap septum, hair, whorl scar, interwhorl and rhizoid) in the GC analysis). The NMR studies should confirm this. contained predominantly mannan and the haplophase wall However, a discrepancy in our results exists over the extent (gametangia) contained predominantly cellulose (Table 3). of crystallinity of the mannan. Whereas the NMR analysis Our compositional analyses agree with the results of reported that little more than 10% of the mannan was Zetsche et al. (1970), who reported that there is more crystalline (Table 2), almost all of the mannan was resistant galactose than rhamnose in the gametophore, interwhorl to acid hydrolysis and hence expected to be crystalline and rhizoid walls of A. mediterranea, except the hairs for (Table 3). It is possible that the NMR underestimated the which we reported that rhamnose exceeded galactose. amount of crystallinity insofar as the interpretation of this Furthermore, Zetsche et al. (1970) detected xylose in kind of data has been developed principally from cellulose, gametangial walls only, whereas we found xylose, as well and some feature of the mannan, or indeed of the as arabinose, in all seven regions, albeit in small amounts A. acetabulum cell wall, may interfere. On the other hand, (Table 3). Whorl scars and rhizoids had a similar ratio of even mild acid hydrolysis, such as occurs during oxalate trace sugars, which differed from the ratio of trace sugars extraction (MacKay et al. 1988), can lead to an artifactual that seemed to be common among the interwhorls, crystallization by removing side chain sugars followed by an gametophores and cap septa. It is noteworthy that the ordering of the remaining long chains into a functional most morphologically complex regions, whorl scar and crystal. In view of the fact that the NMR was done on rhizoid, were simplest in their sugar composition, contain- hydrated cell walls, the NMR data may be more reliable, ing 490% of total recovered sugar as mannose. Hence, and the mannan in muro may be only modestly crystalline. biochemical and morphological complexity are not simply What is more, even the portion of the wall assessed by correlated. NMR as crystalline appeared to be somewhat less well Hairs and septa, both regions of extensive compart- ordered than typical of cellulosic walls. Comparison of the mentalization and segmentation, had the lowest percentage
T1D for the A. acetabulum crystalline mannan component of mannose, 82%. Although delicate and ephemeral, (10–30 ms) with that for cotton and bean cellulose the hairs comprise half the volume of a vegetative thallus (30–60 ms) suggests that the packing of polymers into (Ngo et al. 2004). However, the walls of hairs are not as 130 Cell walls of Acetabularia thick as that of the interwhorl, rhizoid or gametophore Popper and Fry (2003) showed that C. corallina lacked (von Dassow et al. 2001; D. Capps and D.F. Mandoli, isoprimeverose, the dissacharide containing terminal xylose unpublished data) and so probably represent 550% of the [t-Xyl-(1!6)-a-D-Glc] and diagnostic for xyloglucan, organism’s total cell wall. The amount of rhamnose in and instead contained small amounts of 4-linked xylan. various wall regions is roughly correlated with the degree of In A. acetabulum, while the GlcA found could account for flexibility of that anatomical region. The proportion of some of the side groups at the branch point xylose residues, rhamnose was greatest in hairs, followed by gametophores the amount of GlcA is uncorrelated with xylan content, and cap septa, with the least in rhizoids and whorl scars making the GlcA more likely to be a constituent of another (Table 4). This order matches the regional flexibility in the polymer. The roles of each of these non-mannan, diploid thallus: hairs are the most flexible region, often non-cellulosic polysaccharides and their relative abundance Downloaded from https://academic.oup.com/pcp/article/48/1/122/2469303 by guest on 29 September 2021 undulating in the moving seawater, whereas rhizoids are in each of the wall regions remains to be determined. stiff and typically motionless. Physical measures of wall regions that differ in rhamnose content are needed to Strength through weakness: implications of the mannan-rich substantiate and quantify these observations. wall The linkage analyses revealed that A. acetabulum The mannan-rich walls of the diplophase are likely to contains polymers related to those typically found in be more extensible than the cellulose-rich walls of other flowering plants, albeit at much lower amounts compared algae or higher plants. This follows from the NMR analysis with the mannans and cellulose. Because of the presence of showing that the polymers from diplophase cell walls have 4-linked GalA and 2-linked rhamnose, it is tempting to limited crystallinity. Compared with land, water is a suggest that A. acetabulum contains a primitive form of buoyant environment, meaning the cell wall of marine rhamnogalacturonan I (RG I). Indeed, this type of acidic organisms can be less rigid than it would have to be on land. polysaccharide is known to be present in green algae, A flexible wall could be suitable for dealing with wave insofar as antibodies against plant RG I recognize an action and other intermittent stresses in the ocean. epitope in C. corallina cell walls (Wang-Cahill and Kiss Mechanical tests to our knowledge have not been 1995), and a heteropolymer containing rhamnosyl and performed on walls from this alga, and would be useful to glucuronosyl residues has been purified from an Ulva confirm this point. species (Lahaye and Ray 1996). However, charophyte cell In the wild, A. acetabulum thalli are rigidified by means walls contain uronic acids in considerable quantity of calcification (Serikawa et al. 2000). In addition to (Morrison et al. 1993) whereas in A. acetabulum walls, increasing strength or toughness, calcification may help sugar acids are scarce and the amount of 2-rhamnose prevent predation (Marin and Ros 1992). In walls of algae exceeds that of 4-GalA, sometimes by 20-fold. While the and higher plants with acidic polysaccharides, calcium ions existence of 2-rhamnose and 4-GalA in the same polymer form cross-bridges, and contribute in that way to the needs to be tested, the 2-rhamnose is more likely to strength of the wall. In A. acetabulum, uronic acids are all constitute a sulfated homopolymer of 2-linked rhamnan, but absent; it is conceivable that sulfate residues play a like that found in Chlorophyta species (Lee et al. 1998, similar role. Nevertheless, the amount of calcium accreting Mitova et al. 1999), although our preparation techniques to the wall is large, suggesting that the ion is present within did not account for recovery of sulfate. Additionally, we the wall as precipitates or some bio-mineralized form. were unable to detect 2,4-rhamnose, which indicates that Flexibility of the mannans might facilitate incorporation of any rhamnan is unbranched and hence lacks the neutral calcium clusters of different sizes. (1!5)-a-L-arabinan and (1!4)-b-D-galactan side chains As far as we know, this group of mannan weeds did not attached to the rhamnosyl O-4, branches that constitute a evolve land-dwelling forms, while cellulosic orders, such as conserved characteristic of flowering plant RG I (Carpita the Charales, did. It is possible that using a flexible mannan and Gibeaut 1993). wall for vegetative growth and the more rigid cellulose wall The cap and rhizoids contained significant amounts of for reproduction has had implications for the niche that this 4,6-mannose, which is typical of the galactomannans and species can occupy and for aspects of morphogenesis and glucogalactomannans of flowering plants (Meier and development. When mature gametangia are isolated from Reid 1982). The linkage analysis also suggests that the gametophores manually, they exit most easily by the way A. acetabulum walls contain small amounts of branched that the haploid nuclei came in, toward the center of the cap xylans, similar to those found in flowering plants. Although even though the gametophores are wider at the periphery. 4- and 4,6-glucose residues typical of xyloglucans were As the rigid, mature gametangia exit, the mannan-rich observed, we found only trace amounts of terminal xylose. gametophore wall flexes in a manner that a more rigid Xyloglucans were once thought to be components of cellulosic wall might not. To release its spores, the ‘inky cap’ the charophyte wall but, consistent with our findings, mushroom, Coprinus cinereus, autolyses its cap by Cell walls of Acetabularia 131 activating glucanases and chitinases (Buller 1909, Ku¨es For measurement, caps or stalks were packed into an NMR tube, 2000). As the cap of A. acetabulum matures, the septal area either as prepared or after exchange with D2O. Second moment, solid echo (M2interpair) and Jeener echo (T1D) measurements were and gametophores seem to become increasingly flexible made using procedures and pulse sequences described previously while the gametangia become increasingly tough and rigid. (MacKay et al. 1988, Taylor et al. 1990, Wallace et al. 1993). It is plausible that activation of a latent enzymatic activity in the cap matrix (the clear gel-like substance that surrounds Dissection of thalli for monosaccharide and methylation analyses the gametangia) preferentially digests the mannan wall of One hundred or more reproductive individuals bearing the parent, thus leaving the gametangia free to disperse. gametangia were dissected into six parts: gametophores, cap septa, gametangia, interwhorls, whorl scars and rhizoids. Whorls That this taxon committed to a mannan wall for vegetative of hairs were obtained from vegetative late adults because whorls growth, while using a cellulose wall only for reproduction, are usually shed during reproduction. The ‘skirt’ of gametophores Downloaded from https://academic.oup.com/pcp/article/48/1/122/2469303 by guest on 29 September 2021 may have prevented the mannan thallus from reaching land, was cut away from its cap septum and cleared of cytoplasm. All much like the exclusive use of mitosis for reproduction dissected parts were boiled in 95% ethanol for 10 min to remove apparently has committed the diplophase of this species any soluble sugars and remaining cytoplasm. Gametangia were to unicellularity. cleared of cytoplasm by homogenization for 30 s with a motorized, sintered glass pestle in a sintered glass tube and subsequent boiling in the ethanol solution. All dissected thallus regions were washed Materials and Methods with water before they were dried in vacuo and ground into a fine powder with a sintered glass grinder. Cell culture Acetabularia acetabulum (L. PC Silva; previously classified as Monosaccharide and linkage analyses A. mediterannea), genetically heterogeneous strain Aa0005 (also Each sample (1–2 mg) was hydrolyzed with 2 M TFA called Ladenburg 17), was used for all experiments. All phases of containing 1 mmol of myo-inositol (internal standard) for 90 min the life cycle were grown according to Serikawa et al. (2000). at 1208C. The insoluble, acid-resistant material of each sample Briefly, algal cultures were grown at 21 28C on open shelves (gametophore, interwhorl and gametangia) remaining after hydro- directly under cool white fluorescent lights on a 14 h light/10 h dark lysis was pelleted by centrifugation, and the amount of cellulose or photoperiod. Material for NMR was grown to 2 cm tall in the mannan was determined by phenol–sulfuric assay (Dubois et al. Mandoli laboratory growth chamber, shipped by courier to 1956) with cellulose and mannose as standards. For the soluble Vancouver, stored frozen at 208C, and then thawed as required. fraction, TFA was evaporated under a stream of nitrogen and the Material for FTIR and linkage analysis was grown to 1–2 cm tall sugars were converted to alditol acetates (Gibeaut and Carpita in the Mandoli laboratory growth chamber. These cells 1991). The alditol acetates were separated by gas–liquid chroma- were maintained at Friday Harbor Laboratories until they were tography on a 0.25 mm 30 m vitreous silica capillary column of dissected and boiled in 70% ethanol. Finally, these cells SP-2330 (Supelco, Bellefonte, PA, USA). The temperature was were shipped at room temperature by courier to Purdue. held at 808C for 1 min upon injection then programmed from 80 to 1708Cat258C min 1, then to 2408Cat58C min 1, with a 6 min hold at the upper temperature. Monosaccharide identity, as FTIR microspectroscopy pentose, hexose and deoxysugar, was confirmed by electron Gametangia-bearing reproductive individuals were used. impact MS (Carpita and Shea 1989). Gametophores were separated from cap septa and manually For linkage analysis, polysaccharides were per-O-methylated cleared of gametangia. The gametangia were treated separately with lithium-methylsulfinylmethanide, prepared by addition of from the gametophores and cap septa. Samples of isolated wall n-butyllithium to dry dimethlsulfoxide (DMSO), and methyl regions were spotted on gold-plated reflective slides (EZ-Spot, iodide, according to Gibeaut and Carpita (1991). The per- Spectra-Tech, Shelton, CT, USA) and dried at 378C for about 1 h. O-methylated polymers were recovered after addition of water to The slides were supported on the stage of a Nicolet/Spectra Tech the mixture and partitioning into chloroform. The chloroform ‘Continuum’ series microscope accessory of a Nicolet Model 670 extracts were washed five times with a 3-fold excess of water each, FTIR spectrometer equipped with a liquid nitrogen-cooled and the chloroform was evaporated in a stream of nitrogen gas. MCT-A detector. An area of wall (150 mm 150 mm) was selected The methylated polymers were purified by chloroform partitioning for spectral collection in reflectance mode. Interferograms were 1 and hydrolyzed in 2 M TFA for 90 min at 1208C. The TFA was collected with 8 cm resolution and, to improve the signal- evaporated in a stream of nitrogen gas, and the sugars were then to-noise ratio, 128 of them were co-added for each region of the reduced with NaBD4 and acetylated. The partly methylated alditol wall. To calculate difference spectra, three spectra were collected acetates were separated on the column as the alditol acetates. After from different areas of each representative wall region, a hold at 808C for 1 min during injection and rapid ramping, the baseline-corrected, area-normalized and then subtracted. derivatives were separated in a temperature program of 160–2108C at 28C min 1, then to 2408Cat58C min 1, with a hold of 5 min at Nuclear magnetic resonance the upper temperature. All derivative structures were confirmed by Whole caps from immature reproductive thalli lacking electron impact MS (Carpita and Shea 1989). gametangia were manually dissected from their stalks. Because NMR demands gram quantities of tissue, these regions of the Amino acid, amino sugar and uronic acid composition unicell were not further dissected. Walls were spread between two The limited availability of regional wall material made it glass fiber filters and squeezed to remove free liquid water impossible to complete a full amino or uronic acid analysis of all and subjected to repeated washing. After these treatments, differentiated regions; therefore, we compared amino and uronic contaminating cytoplasm was undetectable by light microscopy. acids in three major regions: vegetative, reproductive and haploid 132 Cell walls of Acetabularia walls for amino acids, and vegetative, gametophore (a specific Dubois, M., Gilles, D.A., Hamilton, J.K., Rebers, P.A. and Smith, F. region of reproductive cells) and haploid walls for uronic acids. (1956) Colorimetric method for the determination of sugars and related Cells were sonicated to loosen cytoplasm, ground in liquid substances. Anal. Chem. 28: 350–356. Filisetti-Cozzi, T.M.C.C. and Carpita, N.C. (1991) Measurement of uronic nitrogen, washed with deionized water until the residual pellet was acids without interference from neutral sugars. Anal. Biochem. 197: free of cytoplasm and green material, and lyophilized. General 157–162. monosaccharide composition was determined by HPLC analysis of Frei, E. and Preston, R.D. (1968) Non-cellulosic structural polysaccharides concentrated and 2 M TFA hydrolysates. Amino acid and amino in algal cell walls. III. Mannan in siphoneous green algae. Proc. R. Soc. sugar composition determination was performed on cytoplasm-free B: Biol. Sci. 169: 127–145. wall preparations of stalks and caps that were hydrolyzed with 6 M Gibeaut, D.M. and Carpita, N.C. (1991) Tracing the biosynthesis of the cell HCl in vacuo for 21 h at 1208C. The hydrolysates were dried in wall in intact cells and plants. Selective turnover and alteration of vacuo, dissolved in 0.1 M HCl and analyzed directly using a cytoplasmic and cell wall polysacchrides of proso millet cells in liquid culture and Zea mays seedlings. Plant Physiol. 97: 551–561.
Beckman 1208C automatic amino acid analyzer operated following Downloaded from https://academic.oup.com/pcp/article/48/1/122/2469303 by guest on 29 September 2021 Graham, L.E. and Wilcox, L.W. (2000) Algae. Prentice Hall, Upper Saddle the manufacturer’s instructions, but with the short column River, NJ. lengthened to 15 cm to allow resolution of glucosamine and Iriki, Y. and Miwa, T. (1960) Chemical nature of the cell wall of the green galactosamine (Taylor 1981). The method of Bergman and Loxley algae, Codium, Acetabularia and Halicoryne. Nature 185: 178–179. (1970) was used to assay hydroxyproline. Kacura´kova´, M., Capek, P., Sasinkova´, V., Wellner, N. and Total uronic acids in samples of the gametophore, stalks Ebringerova´, A. (2000) FT-IR study of plant cell wall model compounds: (whorl scars plus interwhorl, but no rhizoids or hairs) and pectic polysaccharides and hemicelluloses. Carbohydr. Polym. 43: gametangia were determined by colorimetric assay (Filisetti- 195–203. Kim, J.-B. and Carpita, N.C. (1992) Changes in esterification of the uronic Cozzi and Carpita 1991), and compared with total sugar acid groups of cell wall polysaccharides during elongation of maize determined by phenol–sulfuric assay (Dubois et al. 1956). To coleoptiles. Plant Physiol. 98: 646–653. determine the type of uronic acid, wall material was carboxyl- Ku¨es, U. (2000) Life history and developmental processes in reduced with sodium borodeuteride after activation with a water- the basidiomycete Coprinus cinereus. Microbiol. Mol. Biol. Rev. 64: soluble diimide (Carpita and McCann 1996). The former uronic 316–353. acid was determined by electron impact MS according to the Lahaye, M. and Ray, B. (1996) Cell-wall polysaccharides from the marine equation described by Kim and Carpita (1992), which accounts for green alga Ulva ‘rigida’ (Ulvales, Chlorophyta)—NMR analysis of ulvan 13C spillover. polysaccharides. Carbohydr. Res. 283: 161–173. Lee, J.B., Yamagaki, T., Maeda, M. and Nakanishi, H. (1998) Rhamnan sulfate from cell walls of Monostroma latissimum. Phytochemistry 48: 921–925. Acknowledgments Liang, C.Y. and Marchessault, R.H. (1959) Infrared spectra of crystalline polysaccharides II. Native celluloses in the region from 640 to 1700 cm 1. We thank Sheila Mehri and Margaux Moffet for help in J. Polym. Sci. 39: 269–278. dissecting wall regions of distinct morphology in the initial stages MacKay, A.L., Bloom, M., Tepfer, M. and Taylor, I.E.P. (1982) of this project. We owe enormous thanks to Tobias Baskin for A broadline proton magnetic resonance study of cellulose, pectin and bean cell walls. Biopolymers 21: 1521–1534. input on and editing of this manuscript. This work was supported MacKay, A.L., Tepfer, M., Taylor, I.E.P. and Volke, F. (1985) Proton in part by a Friday Harbor Laboratories Apprenticeship 2000 NMR moment and relaxation study of cellulose morphology. (E.D.) and a Mary Gates Undergraduate Research Fellowship Macromolecules 18: 1124–1129. (E.D.), NSERC of Canada operating grants (A.L.M. and I.E.P.T.) MacKay, A.L., Wallace, J.C., Sasaki, K. and Taylor, I.E.P. (1988) and NSF IBN#9630618, MCB#0317151 and DEB#0228655 Investigation of the physical structure of the primary plant cell wall (D.F.M.). Journal paper No. 17240 of the Purdue University by proton magnetic resonance. Biochemistry 27: 1467–1473. Agricultural Experiment Station. Marin, A. and Ros, J.D. (1992) Dynamics of a peculiar plant–herbivore relationship: the photosynthetic ascoglossan Elysia timida and the chlorophycean Acetabularia acetabulum. Mar. Bot. 112: 677–682. References McCann, M.C., Chen, L.-M., Roberts, K., Kemsley, E.K., Se´ne´,C., Carpita, N.C., Stacey, N.J. and Wilson, R.H. (1997) Infrared micro- spectroscopy: sampling heterogeneity in plant cell wall composition and Bergman, I. and Loxley, R. (1970) The determination of hydroxyproline in architecture. Physiol. Plant. 100: 729–738. urine hydrolysates. Clin. Chim. Acta 27: 347–349. Meier, H. and Reid, J.S.G. (1982) Reserve polysaccharides other than Buller, A.H.R. (1909) The Coprinus type of fruit-body. In Researches on starch in higher plants. In Encyclopedia of Plant Physiology. Edited by Fungi. pp. 196–215. Longmans, London. Loewus, F.A. and Tanner, W. pp. 418–471. Springer-Verlag, Berlin. Carpita, N.C. and Shea, E.M. (1989) Linkage structure by gas Mitova, M.I., Usov, A.I., Bilan, M.I., Stefanov, K.L., Dimitrova- chromatography–mass spectrometry of partially-methylated alditol Konaklieva, S.D., Tonov, D.P. and Popov, S.S. (1999) Sterols and acetates. In Analysis of Carbohydrates by GLC and MS. Edited by polysaccharides in freshwater algae Spirogyra and Mougeotia. Beermann, C.J. and McGinnis, G.D. pp. 155–216. CRC Press, Boca Z. Naturforsch. 54: 1016–1020. Raton, FL. Ngo, D.A., Garland, P.A. and Mandoli, D.F. (2004) Development and Carpita, N.C. and Gibeaut, D.M. (1993) Structural models of primary cell compartmentation of the central vacuole of Acetabularia acetabulum. walls in flowering plants: consistency of molecular structure with the New Phytol. 165: 731–746. physical properties of the walls during growth. Plant J. 3: 1–30. Nishimura, N.J. and Mandoli, D.F. (1992a) Population analusis of Carpita, N.C. and McCann, M.C. (1996) Some new methods to study reproductve cell structures of Acetabularia acetabulum (Chlorophyta). plant polyuronic acids and their esters. In Progress in Glycobiology. Phycologia 31: 351–358. Edited by Townsend, R. and Hotchkiss, A. pp. 595–611. Marcel Dekker, Nishimura, N.J. and Mandoli, D.F. (1992b) Vegetative growth of New York. Acetabularia acetabulum (Chlorophyta): structural evidence for juvenile Chanzy, H.D., Grosrenaud, A., Vuong, R. and Mackie, W. (1984) The and adult phases of development. J. Phycol. 28: 667–677. crystalline polymorphism of mannan in plant cell walls and after Popper, Z.A. and Fry, S.C. (2003) Primary cell wall composition of recrystallization. Planta 161: 320–329. bryophytes and charophytes. Ann. Bot. 91: 1–12. Cell walls of Acetabularia 133
Proseus, T.E. and Boyer, J.S. (2006) Calcium pectate chemistry controls Tsuboi, M. (1957) Infrared spectrum and crystal structure of cellulose. growth rate of Chara corallina. J. Exp. Bot. 57: 3989–4002. J. Polym. Sci. 25: 159–171. Serikawa, K.A., Porterfield, D.M., Smith, P.J.S. and Mandoli, D.F. (2000) von Dassow, M., Odell, G. and Mandoli, D.F. (2001) Relationships Calcification and measurements of net proton and oxygen flux reveal between growth, morphology and wall stress in the stalk of Acetabularia subcellular domains in Acetabularia acetabulum. Planta 211: 474–483. acetabulum. Planta 213: 659–666. Taylor, I.E.P. (1981) An improved separation of glucosamine in routine Wallace, J.C., MacKay, A.L., Sasaki, K. and Taylor, I.E.P. (1993) A proton amino acid analysis of plant glycoproteins. Phytochemistry 20: nuclear magnetic resonance study of the physical changes in growing 2769–2770. plant cell walls. Planta 190: 227–230. Taylor, I.E.P. and Cameron, D.S. (1973) Preparation and quantitative Wang-Cahill, F. and Kiss, J.Z. (1995) The statolith compartment in analysis of fungal cell walls: strategy and tactics. Annu. Rev. Microbiol. Chara rhizoids contains carbohydrate and protein. Amer. J. Bot. 82: 27: 243–259. 220–229. Taylor, I.E.P., Wallace, J.C., MacKay, A.L. and Volke, F. (1990) Use of Zetsche, K., Grieninger, G.E. and Anders, J. (1970) Regulation of enzyme chemical fractionation and proton nuclear magnetic resonance to probe activity during morphogenesis of nucleate and anucleate cells of
the physical structure of the primary plant cell wall. Plant Physiol. 94: Acetabularia. In Biology of Acetabularia. Edited by Brachet, J. and Downloaded from https://academic.oup.com/pcp/article/48/1/122/2469303 by guest on 29 September 2021 174–178. Bonotto, S. pp. 87–110. Academic Press, New York.
(Received September 5, 2006; Accepted November 21, 2006)