Identification of EDTA-Soluble Polysaccharides from Pea Epicotyl Cell Walls and Their Interaction with Xyloglucan

doi:10.4236/ajps.2011.22016

Paper Menu >>

Journal Menu >>

American Journal of Plant Sciences, 2011, 2, 148-155

doi:10.4236/ajps.2011.22016 Published Online June 2011 (http://www.SciRP.org/journal/ajps)

Identification of EDTA-Soluble Polysaccharides

from Pea Epicotyl Cell Walls and Their

Interaction with Xyloglucan

Elias A.-H. Baydoun1, Oula C. Mansour1, Sandra E. Rizk2, Christopher T. Brett3

1Department of Biology, American University of Beirut, Beirut, Lebanon; 2Department of Natural Sciences, Lebanese American

University, Beirut, Lebanon; 3Institute of Biomedical Life Sciences, University of Glasgow, Glasgow, United Kingdom.

Email: [email protected], oula[email protected], sandra.rizk@lau.edu.lb, [email protected]

Received December 9th, 2010; revised March 23rd, 2011; accepted March 31st, 2011.

ABSTRACT

Nascent pectin and glucuronoxylan were prepared from membrane-bound enzymes obtained from pea epicotyls. They

had previously been shown to exhibit a protein- and pH-dependent pattern of binding to cell wall ghosts and to xy-

loglucan extracted from cell walls prepared from pea epicotyls; maximum binding required a pH of 3-4, and the pres-

ence of cell wall proteins, namely assemblins. To determine whether wall polysaccharides deposited in cell walls be-

have in the same manner as nascent polymers, radioactively labeled EDTA-soluble polymers were prepared from

newly-deposited pea epicotyl cell walls. Different enzyme treatments followed by column chromatography, in addition

to complete acid hydrolysis followed by paper and thin layer chromatography, indicated the presence of pectin, to-

gether with smaller amounts of glucuronoxylan, in this EDTA-soluble extract. These radioactively labeled polysaccha-

rides were found to bind to cell wall ghosts and to xyloglucan extracted from the second and third internodes of pea

epicotyls cell walls in a pH-dependent manner, similar to the binding pattern obtained with nascent polymers. Maxi-

mum bindi ng occurre d at pH 3-4, and also r e qu ired the pr esence of p ro tein.

Keywords: Assemblin, Cell Wall Assembly, Glucuronoxylan, Pectin, Pisum Sativum, Xyloglucan

1. Introduction

The plant cell wall contains cellulose microfibrils em-

bedded in a non-crystalline matrix. Matrix polysaccha-

rides are synthesized in the Golgi apparatus and trans-

ported in vesicles to the wall [1]. There, the newly syn-

thesized matrix polysaccharides associate with each other

and with newly synthesized cellulose to form the new

innermost layer of the cell wall.

Xyloglucan, the principal hemicellulose of the primary

cell walls of dicotyledonous plants, plays an essential

role in cell wall elongation due to its strong association

to cellulose microfibrils [2,3]. Nascent EDTA-soluble

14C-polysaccharides prepared from pea seedlings (Pisum

sativum) consist of radioactively labeled pectins and glu-

curonarabinoxylans (GAX), with a significantly greater

abundance of pectins [4]. These polysaccharides bound

to xyloglucan in a pH-dependant manner, with the high-

est binding occurring at pH 3 - 4, which corresponds to

the pH of a growing wall. The binding decreased to al-

most zero at pH 6 [5]. This binding pattern indicates a

significant role in cell-wall assembly during deposition

and cell wall extension during growth. Since

pre-treatment of nascent GAX with protease greatly de-

creased the binding at acidic pH and abolished the

pH-dependant binding pattern, the protein attached to

nascent GAX, named assemblin, appeared to have an

essential role in the deposition of GAX into the cell wall

[5]. However, cell wall polymers are synthesized in the

Golgi apparatus and undergo several modifications prior

to their excretion into the cell wall space [6]. The present

study aims to determine whether newly-deposited poly-

saccharides extracted from preformed cell walls behave

in the same manner as nascent polymers, with respect to

their interaction with cell-wall ghosts and with xyloglu-

can.

2. Materials and Methods

2.1. Plant Material

Peas (Pisum sativum L.cv Meteor, obtained form Sharpes

International, Sleaford, UK) were soaked overnight at

Identification of EDTA-Soluble Polysaccharides from Pea Epicotyl Cell Walls and 149

Their Interaction with Xyloglucan

room temperature and grown on damp vermiculite at 25˚C

in continuous darkness. For xyloglucan extraction, the

peas were grown for 8 - 9 days. For the preparation of

cell wall ghosts, nascent cell wall polymers and

newly-deposited cell wall polymers, the growth period

was 6 days.

2.2. Preparation of Cell Wall Ghosts and

Extraction of Xyloglucan

A modification of the methods of Hayashi and Maclach-

lan (1984) and Ogawa et al. (1990) was used [7,8]. Third

internodes (50 g) were harvested from 8-9-day-old peas

and extracted three times with 70% (v/v) ethanol (150 ml,

30 min, 70˚C). The tissue was chopped with a razor

blade, homogenized using a pestle and mortar in

Tris-HCl buffer (0.1 M, pH 7.0, 75 ml), and centrifuged

(8000 g, 10 min). The cell wall pellets were extracted

three times with EDTA buffer (0.1 M, pH 7.0, 75 ml, 30

min, 85˚C), and three times with 0.71 M KOH/26 mM

NaBH4 (75 ml, 1 h, 25˚C) in a shaking incubator. Insolu-

ble material (cell wall ghosts) was obtained by centrifu-

gation (8000 g, 10 min). To prepare xyloglucan, the cell

wall ghosts were extracted twice with 4.3 M KOH/26

mM NaBH4 (25 ml, 4 h, 25˚C) in the same incubator.

The combined 4.3 M KOH-soluble fractions were neu-

tralized with acetic acid. Ethanol was added 70% (v/v),

and xyloglucan allowed to precipitate overnight. The

precipitated material was treated with salivary amylase

(24 h, 40˚C) and protease (Type XIV, Sigma, 24 h, 37˚C).

Xyloglucan content was determined by iodine-sodium

sulphate method [9,10].

2.3. Preparation of Nascent EDTA-Soluble

14C-Polysaccharides

The procedure for particulate enzyme preparation was

similar to that of Waldron and Brett (1987) and Rizk et al.

(2000) [5,11]. Epicotyls (6 - 9 cm long) were cut off, the

hooks discarded and the remaining tissue cooled on ice.

Subsequent operations were carried out at 0˚C - 4˚C.

Epicotyls were homogenized using a pestle and mortar in

Tris-Mes buffer (10 mM, pH 6.0). The homogenate was

strained through two layers of muslin, and the filtrate

centrifuged at 100,000 g for 30 minutes. The pellet was

homogenized in cold homogenization buffer using a

glass-Teflon tissue homogenizer, to obtain the particulate

enzyme preparation, which was then incubated (25˚C, 4

h) with UDP-D-[U-14C]glucuronic acid (3.53 kBq, 1.7

μM), UDP-xylose (1 mM) and MnCl2 (10 mM). The re-

action was terminated by adding 96% (v/v) ethanol. After

centrifugation (10,000 x g, 5 min), particulate material

was washed three times with 70% (v/v) ethanol and ex-

tracted twice with 50 mM EDTA/50 mM sodium phos-

phate buffer (pH 6.8) for 5 min at 100˚C. The combined

EDTA/phosphate extracts were passed through a Sephadex

G-100 column, eluted with water, and the excluded

material was collected for binding experiments [5].

2.4. Preparation of Newly-Deposited Cell Wall

EDTA-Soluble 14C-Polysaccharides

Newly-deposited cell wall 14C-polysaccharides were pre-

pared by incubating pea epicotyls with [14C]sucrose

(0.148 MBq at the base of 10 epicotyls) for 24 hours.

Cell wall pellets were obtained from these epicotyls as

described above, and then extracted with boiling 50 mM

EDTA/50 mM sodium phosphate buffer (pH 6.8) for 5

minutes. The EDTA-extracts obtained were passed through

a column of Sephadex G-100 and eluted with water to

remove low molecular weight products. The high mo-

lecular weight material which ran parallel to blue dextran

was combined and used for analysis and binding assays

to xyloglucan.

2.5. Assay of Binding of 14C-Polysaccharides to

Xyloglucan or to Cell Wall Ghosts

Pea xyloglucans or cell wall ghosts (1 mg/incubation)

were resuspended in 0.5 ml incubation buffer (10 mM

oxalate/10 mM phosphate, adjusted to the appropriate pH

with HCl or NaOH), mixed with 0.5 ml 14C-polysaccha-

rides (generally 15 - 30 Bq), incubated (5 min, 25˚C and

centrifuged (5 min, 10,000 x g). Pellets were washed

once with buffer of the appropriate pH (0.5 ml), centri-

fuged for 5 min, resuspended in 0.4 ml water and mixed

with 4 ml Ultima-Flo AF (Packard Instrument Company,

Meriden, Conn., USA) for liquid scintillation counting.

All binding experiments were carried out in duplicate.

Results are expressed as the mean ± the difference be-

tween the experimental values and the means.

2.6. Acid Hydrolysis

Newly-deposited cell wall 14C-polysaccharides were acid

hydrolyzed using trifluoroacetic acid (2 M, 120˚C, 1 h).

The hydrolysate was centrifuged at 10,000 x g and the

supernatant was rotoevaporated to be analyzed by paper

and thin layer chromatography.

2.7. Chromatography

Paper chromatography (PC) was performed using

Whatman no. 3 paper, run in a solvent of ethyl acetate:

pyridine: water (8:2:1 v/v) for 24 hours. Thin layer

chromatography (TLC) was performed using silica coated

plates and run in a solvent of butan-1-ol: ethyl acetate:

water (7:1:2 v/v) for 18 - 22 hours [12]. 1 cm strips (for

the PC) and 0.5 cm strips (for the TLC) were cut from

the chromatogram for scintillation counting.

Identification of EDTA-Soluble Polysaccharides from Pea Epicotyl Cell Walls and

Their Interaction with Xyloglucan

150

2.8. Enzyme Treatments

Radioactively labeled polysaccharides were subjected to

several enzyme treatments prior to analysis by gel chro-

matography. Conditions used for enzyme treatments

were as follows: Pectin lyase from Aspergillus japonicas

(Sigma): 13 units in sodium acetate (pH 4) at 40˚C for 16

h; xylanase from Trichoderma viride (Megazyme, 40

units) and galactanase from Aspergillus niger (Megazyme,

15 units) in sodium acetate, pH 4.5 at 40˚C for 24 h;

endo-1, 4-β-glucanase (cellulase) from Trichoderma longi-

brachiatum (Megazyme): 5 units in sodium acetate, pH 5

at 50˚C for 16 h. Reactions were stopped by boiling for

10 minutes.

To verify the role of proteins in the interaction be-

tween xyloglucan and newly-deposited EDTA-soluble

14C-polysaccharides, radioactively labeled polysaccha-

rides were treated with 5 units of proteinase K (from

Tritirachium album, Sigma) in Mes, (pH 7) at 37˚C for

16 h prior to performing binding assays. Enzyme treat-

ment was terminated by boiling for 10 minutes.

3. Results

3.1. Identification of EDTA-Soluble

14C-Polysaccharides Prepared from Pea

Stems Incubated with [14C]Sucrose

To determine the nature of the newly-deposited high

molecular weight EDTA-soluble 14C-polysaccharides,

prepared from pea stems incubated with [14C]sucrose,

they were first subjected to complete acid hydrolysis (2N

TFA, 120˚C, 1 h) followed by PC and TLC. Paper chro-

matography showed the presence of radioactively labeled

glucose, with smaller amounts of galactose and minor

amounts of mannose, arabinose, fucose and xylose (Fig-

ure 1). However uronic acids did not migrate under the

experimental conditions used for the paper chromatog-

raphy, so the acid hydrolysate was analyzed by TLC us-

ing silica-coated plates. This indicated the presence of

both galacturonic acid (present in pectins) and glucuronic

acid (present in xylans), in a ratio of 2.5:1, indicating the

greater abundance of pectins than xylans in the extract

(Figure 2).

To confirm this, the EDTA-soluble 14C-polysaccharides

were subjected to several enzyme treatments and the

products were analyzed by column chromatography.

Upon treatment with amylase, the amount of high mo-

lecular weight material did not decrease significantly,

indicating that the presence of glucose revealed by PC

analysis is not due to the presence of starch (Figure 3).

Upon treatment with pectin lyase and passage through

Biogel P2, the amount of high molecular weight material

only decreased slightly (Figure 4); this is probably be-

cause pectin lyase only acts on certain bonds within the

large pectin molecules, producing chains that are large

enough to be excluded from Biogel P2. When using a

Biogel P10 column, which has a higher exclusion limit,

over 50% of the high molecular weight material was

shown to be degraded (Figure 5). The radioactively la-

beled polymers also contain a significant amount of xy-

lans, as indicated by the analysis of xylanase treated ex-

Figure 1. PC of total hydrolysate of cell wall EDTA-soluble polysaccharides. Marker sugars: galactose (7 - 8), glucose (9 - 10),

mannose (12 - 14), arabinose (16 - 18), fucose + xylose (20 - 22), rhamnose (36 - 38).

Identification of EDTA-Soluble Polysaccharides from Pea Epicotyl Cell Walls and 151

Their Interaction with Xyloglucan

Figure 2. TLC of total hydrolysate of cell wall EDTA-soluble polysaccharides. Marker sugars: galacturonic acid (4 - 5), glu-

curonic acid (7 - 9).

Figure 3. Gel filtration on Biogel P2 of amylase products of cell wall EDTA-soluble polysaccharides. BD (7 - 8), CoCl2 (14 - 16).

tracts that were passed through Biogel P2 and P10 col-

-umns: in both cases, 35% - 40% of the radioactivity was

collected as low molecular weight products. The re-

maining polymers that were stable to pectin lyase and

xylanase action were broken down by galactanase and

β(1-4)glucanase, indicating the presence of minor

amounts of galactans and xyloglucan in the extracts

(Figure 6). The presence of xyloglucan in the EDTA-

soluble fraction may be due to covalent bonding to pectin,

esulting in its co-extraction with pectin [13].

Identification of EDTA-Soluble Polysaccharides from Pea Epicotyl Cell Walls and

152

Their Interaction with Xyloglucan

Figure 4. Gel filtration on Biogel P2 of xylanase and pectin lyase products of cell wall EDTA-soluble poly saccharides. BD (6 -

9), CoCl2 (15 - 19).

Figure 5. Gel filtration on Biogel P10 of xylanase and pectin lyase products of cell wall EDTA-soluble polysaccharides. BD (6 - 8),

CoCl2 (14 - 19).

Figure 6. Gel filtration on Biogel P10 of cellulose and galactanase products of cell wall EDTA-soluble polysaccharides. BD (6 -

), CoCl2 (15 - 19). 8

Identification of EDTA-Soluble Polysaccharides from Pea Epicotyl Cell Walls and 153

Their Interaction with Xyloglucan

3.2. Binding of Newly-Deposited EDTA-Soluble

14C-Poly- Saccharides and Nascent

EDTA-Soluble 14C-Polysaccharides to Cell

Wall Ghosts.

Samples of the newly-deposited EDTA-soluble 14C-

polysaccharides were incubated with cell-wall “ghosts”,

i.e. cell wall fragments which had been pre-extracted

with EDTA to remove pectin. Samples of nascent

EDTA-soluble 14C-polysaccharides (prepared by incu-

bating a Golgi-rich membrane fraction with UDP-D-[U-14C]

glucuronic acid) were incubated with cell-wall ghosts in

the same way. Results obtained showed that, over the pH

range 3 to 6, binding of newly-deposited EDTA-soluble

14C-polysaccharides to cell-wall ghosts occurred maxi-

mally at pH 3, which corresponds to the pH of a growing

cell wall [14,15], and decreased with increasing pH until,

at pH 6, it was only 25% of the value at pH 3 (Figure 7).

The pattern of binding of nascent EDTA-soluble 14C-

polysaccharides to cell-wall ghosts was almost the same,

except that the residual binding at pH6 was even lower

(Figure 7).

3.3. Binding of Newly-Deposited and Nascent

EDTA- Soluble 14C-Polysaccharides to

Xyloglucan.

Xyloglucan was extracted and purified from the third

internode of 10-day old pea stems. The binding of

newly-deposited and nascent EDTA-soluble 14C-poly-

saccharides to this xyloglucan over the pH range 3 to 6

was now investigated. The binding patterns obtained

(Figure 8) were very similar both to each other and to

the patterns of binding to cell-wall ghosts (Figure 7). For

nascent EDTA-soluble 14C-polysaccharides, this confirms

the pattern of binding to xyloglucan reported by Rizk

et al. (2000) [5].

Effect of protease treatment of newly-deposited EDTA-

soluble 14C-polysaccharides on binding to xyloglucan.

Figure 7. Effect of pH on binding of nascent and cell wall

EDTA-soluble polymers to cell wall ghosts.

Figure 8. Effect of pH on binding of nascent and cell wall

EDTA-soluble polysaccharides to xyloglucan.

Finally, to check whether newly-deposited EDTA-

soluble 14C-polysaccharides are still linked to protein

after deposition in the cell wall [16], and to verify the

role of this protein in the binding to xyloglucan, the

newly-deposited EDTA-soluble 14C-polysaccharides were

subjected to protease treatment prior to performing the

binding assays over the pH range 3 to 6. The protease

treatment greatly decreased the binding at pH 3 and at

higher pH values (Figure 9). This effect of protease is

very similar to that already reported for the effect of pro-

tease on the binding of nascent EDTA-soluble 14C-poly-

saccharides to xyloglucan [5]. The result confirms that-

protein is required for the binding of newly-deposited

EDTA-soluble wall polysaccharides to xyloglucan after

deposition in the cell wall.

4. Discussion

Nascent GAX was previously reported to bind to hemi-

cellulose from pea epicotyls in a pH-dependent manner,

with the highest binding at pH 3.5 - 4.0; the binding was

thought to occur via non-covalent bonds [4]. Further

Figure 9. Effect of protease pre-treatment of cell wall EDTA-

soluble polysaccharides on the binding to xyloguc a n.

Identification of EDTA-Soluble Polysaccharides from Pea Epicotyl Cell Walls and

154

Their Interaction with Xyloglucan

studies indicated that a similar binding pattern occurs

between nascent GAX and pectin to xyloglucan extracted

from the third internodes of pea epicotyls, and that the

binding requires the presence of proteins named assem-

blins [5]. The association between the polymers would

probably not occur in Golgi vesicles prior to their arrival

at the plasma membrane, since the pH within Golgi vesi-

cles is thought to be close to neutrality. The present in-

vestigation aimed at investigating the interaction between

xyloglucan and polymers that have already been depos-

ited into the wall. The results obtained confirm that the

binding pattern between xyloglucan and EDTA-soluble

polysaccharides that have been extracted from the wall,

is similar to that reported between xyloglucan and nas-

cent EDTA-soluble polysaccharides obtained from Golgi

membranes [5]. To identify these polysaccharides, dif-

ferent enzyme treatments followed by gel filtration were

performed, in addition to acid hydrolysis followed by

chromatography of the hydrlolysate. The results indi-

cated the newly-deposited polysaccharides consisted of

pectin, probably with some xyloglucan attached, and

smaller amounts of glucuronoxylan. These polysaccha-

rides were found to bind to xyloglucan in a pH-depend-

ent manner, whereby pH 3 gave the highest binding over

the pH range tested, which is similar to that reported by

Rizk et al., 2000. This pH corresponds to that of a cell

wall undergoing growth, and lower pH values were not

tested because they would be of doubtful physiological

significance. The binding greatly decreased when the

polysaccharides were subjected to protease treatment

prior to performing the binding assay, confirming the

presence of proteins associated with these polymers after

their deposition into the wall.

This protein- and pH-dependent binding suggests a

functional interaction with the mechanisms that control

growth. In pea epicotyls, the pH of the wall decreases to

3 when growth is initiated [14]. Rapid growth may re-

quire strong interactions between matrix polymers in

order to maintain the cohesion of the wall. In addition,

pectin and GAX molecules present in the cell wall may

compete with cellulose for binding to xyloglucan at an

acidic pH, interfering with the strong xylogluca-

n-cellulose binding, and hence rendering the cell wall

more extensible to allow rapid growth.

REFERENCES

[1] A. Staehelin and I. Moore, “The Plant Golgi Apparatus:

Structure, Functional Organization and Trafficking Mecha-

nisms,” Annual Review of Plant Physiology and Plant

Molecular Bi ology, Vol. 46, 1995, pp. 261-288.

doi:10.1146/annurev.pp.46.060195.001401

[2] S. C. Fry, “Cellulases, Hemicelluloses and Auxinstimu-

lated Growth: A Possible Relationship,” Plant Physiology,

Vol. 75, No. 4, 1989, pp. 532-536.

doi:10.1111/j.1399-3054.1989.tb05620.x

[3] T. Hayashi, K. Ogawa and Y. Mitsuishi, “Characteriza-

tion of the Adsorption of Xyloglucan to Cellulose,” Plant

and Cell Physiology, Vol. 35, No. 8, 1994, pp. 1199-

1205.

[4] C. T. Brett, S. A. Healy, M. S. McDonald, C. Macgregor

and E. Baydoun, “Binding of Nascent Glucuronoxylan to

the Cell Walls of Pea Seedlings,” International Journal of

Biological Macromolecules, Vol. 21, No. 1-2, 1997, pp.

169-173. doi:10.1016/S0141-8130(97)00057-3

[5] S. Rizk, M. R. Abdel, E. Baydoun and C. T. Brett, “Pro-

tein- and pH-Dependent Binding of Nascent Glucurona-

rabinoxylan to Xyloglucan in Pea,” Planta, Vol. 211,

2000, pp. 423-429. doi:10.1007/s004250000303

[6] W. D. Reiter, “Biosynthesis and Properties of the Plant

Cell Wall,” Current Opinion in Plant Biology, Vol. 5, No.

6, 2002, pp. 536-542.

doi:10.1016/S1369-5266(02)00306-0

[7] T. Hayashi and G. Maclachlan, “Pea Xyloglucan and

Cellulose. I. Macromolecular Organization,” Plant Physi-

ology, Vol. 75, 1984, pp. 596-604.

doi:10.1104/pp.75.3.596

[8] K. Ogawa, T. Hayashi and K. Okamura, “Conformational

Analysis of Xyloglucan,” International Journal of Bio-

logical Macromolecules, Vol. 12, No. 3, 1990, pp. 218-

222. doi:10.1016/0141-8130(90)90036-A

[9] P. Kooiman, “A Method for Determination of Amyloid in

Plant Seeds,” Recueil des Travaux Chimiques des

Pays-Bas et de la Belgique, Vol. 79, 1960, pp. 675-678.

[10] T. Hayashi, Y. Kato and K. Matsuda, “Xyloglucan from

Suspension-Cultured Soybean Cells,” Plant and Cell

Physiology, Vol. 21, No. 8, 1980, pp. 1405-1418.

[11] K. W. Waldron and C. T. Brett, “Subcellular Localization

of a Glucuronyltransferase Involved in Glucuronoxylan

Biosynthesis in Pea (Pisum Sativum) Epicotyls,” Plant

Science, Vol. 49, No. 1, 1986, pp. 1-8.

doi:10.1016/0168-9452(87)90013-6

[12] S. C. Fry, “The Growing Plant Cell Wall: Chemical and

Metabolic Analysis,” John Wiley and Sons, New York,

1988.

[13] C. M. Cumming, H. D. Rizkallah, K. McKendrick, R. M.

Abdel-Massih, E. Baydoun and C. T. Brett, “Biosynthesis

and Cell-Wall Deposition of a Pectin-Xyloglucan Com-

plex in Pea,” Planta, Vol. 222, 2005, pp. 546-555.

doi:10.1007/s00425-005-1560-2

[14] R. E. Cleland, G. Buckley, S. Nowbar, N. M. Lew, C.

Stienmetz, M. L. Evans and D. L. Rayle, “The pH Profile

for Acid-Induced Elongation of Coleoptile and Epicotyl

Sections Is Consistent with the Acid-Growth Theory,”

Planta, Vol. 186, No. 1, 1991, pp. 70-74.

doi:10.1007/BF00201499

[15] U. Kutschera, “The Current Status of the Acid Growth

Theory,” New Phytologist, Vol. 126, No. 4, 1994, pp.

549-569. doi:10.1111/j.1469-8137.1994.tb02951.x

Identification of EDTA-Soluble Polysaccharides from Pea Epicotyl Cell Walls and

Their Interaction with Xyloglucan

155

[16] A. Baydoun, M. C. Hobbs, M. H. P. Delarge, M. J. Farmer,

K. W. Waldron and C. T. Brett, “Formation of Glucuron-

oxylan Linked to Protein in Plant Golgi and Plasma Mem-

branes,” Biochemical Society Transactions, Vol. 19, No.

3, 1991, p. 250S.