31“.-

 

' Afiosffigfifnnfi ~ h

r‘m

154k .r

 

 

 

9v -4
”cu..-

.
‘w:

rm

 

,.-'II

.J

 

:3". "‘7 '1

it"

 

 

é
f

5n»

.,._.k
. - -.
'4‘

 

W
‘.

   
 
  

9"“...554' - -’ ‘ . ,
3.1:”.- "f. , _' ‘ ‘ V . "‘ . "“‘>-'"'
. ' u .' n- . will!"

 

unnu- r
, n. "r

   

9.4

11v “1. n

‘x-

..,,
-. r2: - - q ‘ 1 ‘ , . - ’ ""'

,. ..T . .I.‘ ‘ ‘ . .n. ... ‘ \ 4.. v >'-. u "f
shown. A ,1 . v . ;.- ‘ "rm-M T ~-
.. ~ . ' 4.. .9.A.,J. "nu“;
v, . _, . . . r u r“. .. - WV. «\-
l u N ‘ x a n 7 v: " ‘

. .. ,

 

 

run

,1.
n17.1- ~v
4 . 39'“, . q.
-' n .-

 

. \ ' .
(, .,, A._
r ‘ 'zr-sv,

w A...
If." ’ ' ‘
, «#455

.5}. 3:,“
«a

.
,. . ”.5
.. .,
"“"‘~‘:’x'
‘r

3:}... .7

 

» v

13‘. x .

,‘u

tunn-

. .... ,
1

 

J..." . . I.‘ ' -, . , . >
r, - . . , j , ~ -. #41::

"m.- “ ‘7' u 4 ~ - ‘ . u
I \ n - I , u I
p ,' v.“ , .V ‘ lug”
H

v .1 “v I
. , 4,4. , ‘ . . . . v ._
"1"?” ¢ 4...» . - ,1, L‘ ,.‘ ;. .QJ' “a J '. 3'
‘ A ,1 .J , . - ‘ ' Au ‘ v V’ " “
g n s. . - ~;..' ’ l '- r . . ’4' J‘ ;"';x‘,f..:

 

.I
up“) A
an m

4 ‘; v r ' 1.
. t I . _ . .. ‘ ' .1 I ' ... .r
~ . ‘ . . '. .‘ (u [v V ',' .,.,,.,y,v
h H. ,4
.mr ' ' "'
‘1 0"! k '

».

 

.r :9
“Yb!"
«4 ~47- ; A

35;“

 

.' ...,
a.

.A

..

 

...‘~-.;

- “gr--

 

 

 

" ‘4. ‘~ RARIES
l. _b big moms vensnv LIB

M \\\\\\\\\\\\\\\\M\“ H\\\\\2\\\\"\ MW

31

 

 

 

 

 

 

 

 

 

This is to certify that the

thesis entitled

ISOLATION AND CHARACTERIZATION
OF PECTIC POLYSACCHARIDES
FROM Lemna minor

presented by

Liang Chang

has been accepted towards fulfillment
of the requirements for

M 5 degree in BFOW"§+rj

 

 

Major proiessor

Date 5/61/9/

0-7639 MS U is an Affirmative Action/Equal Opportunity Institution

 

 

F LIBRARY 1
Michigan State ‘
University

 

\L

 

—_~_

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due. '

DATE DUE DATE DUE DATE DUE

1 @2321: we

L, 4

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

"1%

MSU Is An Affirmative Action/Equal Opportunity Institution
catamaran]

 

 

 

 

 

ISOLATION AND CHARACTERIZATION
OF PECTIC POLYSACCHARIDES

FROM Lemna minor

By

Liang Cheng

A THESIS
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

MASTER OF SCIENCE

Department of Biochemistry
1991

/ fl
'0‘)

6’2?)- 55

ABSTRACT
ISOLATION AND CHARACTERIZATION OF PEC'I‘IC POLYSACCHARIDES
FROM Lemna minor
By-

Liang Cheng

Pectic polysaccharides were purified from cell walls of Lemna minor and their
homogeneity was examined by column chromatography on DEAE-Sephadex A-25 and
Sephacryl 8-400. The relative molecular size of the samples was determined.

Sugar composition of seven purified pectic polysaccharides were determined by gas
chromatography—mass spectrometry. Galacturonic acid (20.8 to 67.7%) and apiose (23.0
to 39.8%) were the two most abundant components , but rhamnose, fucose, arabinose,
xylose, mannose, glucose and were also found in all samples. Three types of pectic
polysaccharides were found: Samples extracted in 15 min and eluted with 0.25 M NaCl
from DEAE-Sephadex columns have the largest molecular size (FL, of standard dextrans
with the same K“ are 139,000 to 205,000), and are the most acidic (44.8 to 67.7%
galacturonic acid), while those eluted with 0.2 M NaCl have a much smaller molecular
size (N1. of standard dextrans with the same K" are 15,500 to 26,000), high in neutral
sugars (69.8 to 79.2%) and low in galacturonic acid (20.8 to 30.2%). Samples extracted
in a 4.75 h period following the 15 min extraction have the smallest molecular size (M.
of standard dextrans with the same K“, are 6,120 to 112,000) and an intermediate content

of galacturonic acid (34.3 to 38.8%).

To my parents and wife

iii

ACKNOWLEDGMENTS

I would like to thank my major professor, Dr. Paul K. Kindel, for his guidance,
support and patience through these years.

I would like to express my gratitude to my guidance committee, Dr. Rawle I.
Hollingsworth and Dr. Robert S. Bandurski, for their valuable advice.

Special thanks also go to former member of the laboratory, Colleen Curry for her
technical assistance and friendship.

Finally, I am most grateful to my parents, Mo-liang Lu and Yan-ming Cheng, for
their love, dedication and encouragement, and to my wife, Yinghui Dan, for her love,

understanding and support.

iv

TABLE OF CONTENTS

Page
LIST OF TABLES ............................................................................ vii
LIST OF FIGURES ........................................................................... viii
LIST OF ABBREVIATIONS ................................................................. x
INTRODUCTION .............................................................................. 1
Primary Cell Wall, Models ............................................................... 1
Components of the Plant Primary Cell Wall ........................................... 7
Physical and Chemical Properties of Pectic Polysaccharides ....................... 16
PART 1. ISOLATION AND HOMOGENEITY OF THE PECTIC

POLYSACCHARIDES FROM Lemna minor ........................ 24

MATERIALS AND METHODS ............................................................. 24
Materials ..................................................................................... 24
General Methods ............................................................................ 24
Isolation of Pectic Polysaccharides ...................................................... 24
Homogeneity of Pectic Polysaccharides ................................................ 26
RESULTS ........................................................................................ 27
Chromatography on DEAE-Sephadex — Step Gradient ............................. 27
Second Chromatography on DEAE-Sephadex — Linear Gradient ................. 27
Homogeneity of Pectic Polysaccharides ................................................ 32

DISCUSSION .................................................................................... 47
PART 2. SUGAR COMPOSITION OF PECI‘IC POLYSACCHARIDES ............ 49

MATERIALS AND METHODS

Materials ..................................................................................... 49
Reduction of Pectic Polysaccharides .................................................... 49
Analysis of Reduced Pectic Polysaccharides ........................................... 51
RESULTS ......................................................................................... 54
DISCUSSION .................................................................................... 64
LIST OF REFERENCES ...... . ................................................................ 70
APPENDICES
Appendix A. Derivation of Equations for Calculation of the Amount
of Neutral Sugar and Uronic Acid in Samples ....................... 75
Appendix B. Derivation of the Equation Used to Calculate the Percent
of Galactitol Acetate-D2 in Galactitol Acetate ........................ 79
Appendix C. The Acetylation of Apiitol in the Determination of Apiose ......... 83

vi

Table 1.

Table 2.
Table 3.

Table 4.

Table 5.

LIST OF TABLES

Page
Comparison of NaCl Concentrations Needed to Elute Polysaccharide
Samples from DEAE—Sephadex, Relative mobilities, as K“, of Samples
on Sephacryl 8-400, and the Expected Weight Average Molecular
Weight of A Standard Dextran with the Same K“, ............................. 46
Reduction of Cell Wall Pectic Polysaccharides Samples ..................... 55
Sugar Composition of Cell Wall Pectic Polysaccharide Samples ........... 61
Comparison of Selected Properties of the Seven Pectic
Polysaccharide Samples with the Ratio of Rhamnose/
Galacturonic Acid (Rha/GalUA, mole/mice) for These Samples ........... 62
Intensity of Ions in Two Major Ion Clusters from the Mass
Spectrum of Galactitol Acetate/Galactitol Acetate-D2
Prepared from the Pectic Polysaccharides Samples ....................... 63

LIST OF FIGURES

Page

Figure l. Tentative Structure of Sycamore Primary Cell Wall ......................... 3
Figure 2. Extensin-cellulose Network ....................................................... 6
Figure 3. A Summary of the Side Chains Attached to B—4—Linked

D-Xylopyranosyl Residues of the Backbone of Arabinoxylans .............. 10
Figure 4. Structure of Rharnnogalacturonan I .............................................. 13
Figure 5 . Proposed Structures of the Polygalacturonate

Portion of the Pectic Polysaccharide Backbone ................................ 18
Figure 6. Schematic Illustration of the Structure of the Ca2+ Gel Network ......... 21

Figure 7. Column Chromatograms of Cell Wall Pectic Polysaccharide Samples. . . . 29

Figure 8. Column Chromatograms of Rechromatographed
Pectic Polysaccharide Samples ................................................ 31

Figure 9. Column Chromatograms of Pectic Polysaccharides Samples
Obtained from Sample A and Chromatographed Twice
Previously on DEAE-Sephadex ................................................... 34

Figure 10. Column Chromatograms of Pectic Polysaccharide Samples
Obtained from Sample B and Chromatographed Twice
Previously on DEAE-Sephadex ................................................. 36

Figure 11. Comparison of Column Chromatograms Obtained
by the Uronic Acid and Total Sugar Tests ............................ 38

Figure 12. Column Chromatograms of Purified Pectic Polysaccharide
Samples Obtained with Sephacryl S-400 .................................. 41

viii

Figure 13.

Figure 14.

Figure 15.

Figure 16.

Figure 17.

Comparison of Column Chromatograms of

Samples A-lSm—IIIa and A-15m-IIIb ....................................... 43
Relationship between MW of Standard Dextrans and K“ .................. 45
Time Course of Hydrolysis of a Carboxyl—reduced

Pectic Polysaccharide Sample ................................................ 57
Treatment of an Alditol Acetate Sample with Water ....................... 59
Proposed Model for the Primary Cell Wall of L. minor ................... 67

LIST OF ABBREVIATIONS

CMC -—1-Cyclohexy1-3-(2-morpholinoethy1)carbodiimide metho-p—toluenesulfonate
DE —Degree of methylesterifieation

E1 -—Electron impact ionization

GC —Gas chromatography 7

IDT —Isodityrosine

MS —Mass spectrometry

RG-I —Rhamnogalactur"onan I

RG-II —Rhamnogalacturonan II

TFA —Trifluoroacetic acid

INTRODUCTION

Primary Plant Cell Wall, Models

The walls of growing plant cells are ealled primary cell walls. Northcote considered
the primary cell wall to be a growing, constantly changing, composite material consisting
of a dispersed phase of microfibrils within a complex, continuous matrix (57). Lamport
(44), in 1965, considered the primary cell wall to be a single, bag-shaped macromolecule
having a coherent, cross-linked structure, with bonds both between the hydroxyproline-
rich wall protein 'extensin" and wall polysaccharides. Fry (26) gave a simple description
of the primary cell wall, "a biphasic structure, consisting of a rigid skeleton of cellulose
microfibrils held together (and apart) by a gel-like matrix". After investigating the
primary cell wall of suspension-cultured syeamore cells and a number of other plants ('7,
16, 19, 49, 72, 76, 80), Albersheim’s group constructed a model for the primary cell
wall of higher plants (37). This model, which is shown in Figure 1, proposes that the
cellulose microfibrils are linked to each other through xyloglucans, rhamnogalacturonans
(with arabinan and galactan side chains), arabinogalactans and cell wall proteins. In this
model, hemicellulose xyloglucans are tightly held on the surface of the cellulose
microfibrils by hydrogen bonds, the reducing ends of xyloglueans are covalently linked
to galactan side chains of rhamnogalacturonans, which, in turn, are linked to serine
residues of the hydroxyproline-rich cell wall protein through 3,6-linked arabinogalactans.
They suggested that each of the rharnnogalacturonan molecules had a number of galactan

side chains, and that these side chains were connected to a number of different cellulose

Figure l. Tentative Structure of Sycamore Primary Cell Wall (37). [This model is
not intended to be quantitative, but the wall components are presented in approximately
proper proportions, although the distance between cellulose microfibrils is expanded, in
order to allow room to present the interconnecting structures. The key in the Figure
depicts regions representing the various wall components. (a) cellulose microfibrils, (b)
rhamnogalacturonan (pectic backbone), (c) xyloglucan, (d) wall protein with arabinosyl
tetrasaccharides (e) attached to hydroxyl-L-proline residues, (0 (3+6)-linked
arabinogalactan attached to serine (g) of the wall protein, (h) total pectic polysaccharide,
showing L-arabinan and 4-linked D—galactan side-chains attached to rharnnogalacturonan
backbone, and (i) unsubstituted L-serine residues of wall protein] Figure and legend

taken from reference (21).

 

 

 

 

 

 

 

 

 

 

 

O

-r..- o

O.
00

O .4...
fl...

  

 

 

 

 

 

4
microfibrils through xyloglueans. Each cellulose microfibril, in turn, was connected to

several different rhamnogalacturonans through the same type of connection, and in this
way, cellulose microfibrils were extensively cross-linked (37). However, this model and
some other models including Lamport’s early model (42, 45) (which suggests extensin
is linked to cellulose microfibrils by an arabinosyl hydroxy-L—proline-linked galactan, and
intrapeptide and interpeptide linkages of extensin are formed by disulphide bridges) are
now considered to be incorrect bceause they do not explain why, after substantial
cleavage of glycosidic linkages, the extraction of the extensin protein is still difficult.
Based on the fact that after removal of wall polysaccharides through treatment with
hydrogen fluoride the extensin protein remained insoluble (55), and that the isodityrosyl
unit of extensins is found to be an intrapolypeptide bridge and may be an interpeptide
bridges also (25, 27), Lamport proposed a ”warp-weft“ cell wall model (44, 46). This
model, which was based on his earlier dual network hypothesis (43), suggested that the
cross-linked extensin formed a glycoprotein network of defined porosity and that the
cellulose microfibrils penetrated the extensin network through those pores, and connected
with other polysaccharides to form a cellulose based network (Figure 2). A “pH-induced
pectin-extensin switch" hypothesis was suggested based on this model (44). The
hypothesis states that since at low pH the ionization of the carboxyl groups of pectin is
suppressed, the pectin tends to self-associate through carboxyl-carboxyl and carboxyl-
earboxylate hydrogen bonding. Increasing the pH increases the pectin anionic charge,
leading to ion pair formation ("salt bridges") between pectin carboxylate groups and the
protonated e-amino groups of the lysine residues of extensin. In this way a pH change

can alter the rheology of the cell wall as the cell grows. This cell wall model did not

Figure 2. A Hypothetical Extensin-cellulose Network. The perspective view (A)
represents an extensin-rich, i.e. tightly woven primary cell wall of high tensile strength.
A loosely woven wall would contain correspondingly less extensin and would have a
lower tensile strength. The view in cross section (B) shows how two IDT-crosslinked
strands of extensin might define the size and shape of a cellulose microfibril. Figure and

legend taken from reference (46).

EXTENSlN-CELLULOSE NETWORK

A. Perspective view

    
  
 
 

cellulose
microfibril

' PROTOPLAST

plasmamembrane

 

extensin

  
 

isodityroslne
crosslink

CHNH,
co,H

7
describe how polysaccharides are connected with each other. Due to the great technieal

difficulties in isolating and purifying "intact'l polysaccharides from cell walls, the
biochemistry of plant cell walls is still at the stage of identifying and elucidating the

covalent structures of the macromolecular components of the primary cell wall.

Components of the Plant Primary Cell Wall

Primary cell walls of plants are highly hydrated, but on a dry weight basis, are
composed of about 90% polysaccharides and 10% proteins (21, 52) which are mostly
glycoproteins (with small amount of enzymes). Some of the polysaccharides contain
methyl ethers and methyl, acetyl and feruloyl esters (52). Small amounts lipids is also
found in the plant primary cell wall. There are numerous types of polysaccharides in the
primary cell walls of plants and they vary with different plant sources, organs and
different stages of cell growth. Classieally, they are separated into three major groups
(21): The pectic polysaccharides, which ean be extracted by hot water, weak acids and
ammonium oxalate solution and other chelating agents; the hemicelluloses, which ean be
extracted by relatively strong alkali; and the residue remaining, which is mainly cellulose
and extensin proteins.

Cellulose, a polymer of glucose units linked 8-1 ,4, constitutes approximately 20-30%
of the dry weight of the primary cell wall (15, 52, 61). Frequently, small amounts of
other neutral sugars such as xylose and mannose are also found with the cellulose but
whether their presence is the result of contamination remains unclear. Microfibrils, which
are formed by aggregation of chains of cellulose, are the major structural component of

the rigid skeleton ofcell walls.

8
Extensin is an insoluble hydroxyproline-rich (20%) glycoprotein of the primary cell

wall (42, 45, 61, 72). There are about 9 half-isodityrosyl residues per extensin molecule,
and they ean form intrapolypeptide and possibly interpolypeptide isodityrosyl bridges
resulting in a cross-linked extensin molecule network. The cellulose with attached
xyloglueans could be held by the network.

'I\vo major groups of polysaccharides are present in the matrix of the primary cell
walls—hemicellulose and pectic polysaccharides. A typical primary wall of a dicotyledon
contains 15-25% hemicellulose on a dry weight basis (15). Xylogluean is a type of
hemicellulose that consists of a backbone of 3-1 ,4-linked glucosyl residues with a—l ,6-
1inked xylosyl side chains. Some of the xylosyl side chains are extended by the addition
of B-D-Gal» or a-L-Fuc-e2-B-D—Gal-fi to O—2 of the xylosyl residues (7, 20). About 20%
of the primary cell wall of dicots is xylogluean, only 2% for monocots (52). Another
type of hemicellulose is xylan which is the major hemicellulose found in primary cell
wall of monocots, but present only in smaller amounts in dicots. Structurally xylan
consists of a B—l,4-linked xylan backbone with various types of side chains attached to
the O-2 or O-3 position of the xylosyl residues (3, 52). The most common side chain is
a single arabinofuranosyl residue attached to O3 of the xylosyl residues in the backbone,
but different xylans have different side chains and some other examples are shown in
Figure 3. Ferulic and p-coumaric acids have been also found attached to arabinosyl side
chains of xylan through ester bonds (29, 71, 73, 79).

Pectic polysaccharides, or pectic substances, are polysaccharides with a backbone of
galacturonosyl and rhamnosyl residues and having side chains of neutral sugars. The

earboxyl groups of the galacturonosyl residues may be free or partially or firlly methyl

Figure 3. A Summary of the Side Chains Attached to fi-4-Linked D-Xylopyranosyl

Residues of the Backbone of Arabinoxylans. From reference (52).

10

Gal Xyl
la is
L Ara Aga Arz-a
" f f f
l

3 s 3 8 3 a
——\ 4-D-Xyl —> ——,4-D-Xyl —) —-)4-D-Xyl —-)

D-Gl CA 4-O—Me-D-Gl cA Araf

l

2 a 2 a
—; 4-D-Xyl —) —> 4-D—Xyl ——)

2 a
—_) 4-D-Xyl —-)

Gal
is
4

Xyl
is
2

Araf

l

3 a
—) 4-D—Xyl —)

ll
esterified. They are the most abundant matrix component of the primary cell wall and

the middle larnella of most higher plants. They are a complex family of polysaccharides
and make up to 35% of the dry weight of the dicot primary cell wall and (15, 20, 21,
35). Pectic polysaccharides are also widespread in many monocot cell walls (36),
although their contents are usually lower than that in dicots. Traditionally, several neutral
polysaccharides (arabinans, galactans, and arabinogalactans) extracted from cell walls
with the acidic pectic polysaccharides are also ealled pectic polysaccharides. They are
believed to be attached covalently to the backbone primarily through the rhamnosyl
residues (4).

Homogalacturonans are defined as polymers consisting solely or ”predominantly”
(52) of 01-1 ,4-linked-D-galacturonosyl residues. No pure homogalacturonan has been
isolated from primary cell walls without using treatments that are likely to cleave
covalent bonds. However, there is some evidence that they are present in primary cell
walls (18). They have been found in suspension-cultured Rosa cell walls with a degree
of polymerization greater than 100 (17). A polymer containing 91% galacturonosyl
residues was also purified from rice endosperm cells (70), which may suggest the
possible presence of homogalacturonans in monocots. Homogalacturonans have been also
found in the media of tobacco (87% galacturonic acid) (82) and syearnore (95 %
galacturonic acid) (52) cell suspension cultures.

Rhamnogalacturonan I (RG-I), the pectic polysaccharide first isolated from the walls
of suspension-cultured syeamore cells, has been studied in some detail (47, 48, 50, 51,
54). With a degree of polymerization of about 2000, RG-I has a backbone composed of

alternating 2-1inked rhamnosyl and 4-linked galacturonosyl residues (Figure 4A).

12

Figure 4. Structure of Rhamnogalacturonan I. (A) The backbone glycosyl-residue
sequence of rhamnogalacturonan I. From reference (52). (B) A summary of the different
kinds of linking glycosyl residues of the side chains that are attached to 04 of 2,4-1inked

rhamnosyl residues of the backbone. From reference (51).

13

A

side chain side chain
t I

4 4
-) 4-D-GalA 2’ Z-L-Rha 3) t-o-GalA 3 z-t-aha 3» 4-b-GalA 3) Z-L-Rha 3. 4-D-GalA S Z-L-Rha 3)

B
L
.1 3a 4
L-Aras o-Gal D-3Gal boss]
1 J l 4
‘ C ‘ G 4 C ‘ O
—0 24.41113 —0 a Z-L-Rha 4 -) Z-L-Rha —9 —) Z-L-Rha —o

\5 J L l
u-Gal6 o- (Sal6 n-chG
6 J. J.

4 I ‘ C ‘ O
—t Z-L-Rha —o 4 Z—L-Rha -> 4 2-L—Rha —+

14
Approximately half of the 2-linked rharrmosyl residues are branched and at least 30

different side chains, averaging 6-7 residues in length, have been found attached to its
O-4 (50, 52). The first glycosyl residue (attached to O—4 of the rhamnosyl residues of the
backbone) of several different side chains was determined (51) and these are shown in
Figure 4B. The side chains are rich in arabinosyl and galactosyl residues, although small
amounts of fucosyl residues have also been found. Pectic polysaccharides that are similar
to RG-I have been found in other plants, both monocots and dicots, such as tobacco (24),
potato (33), onion (34), maize and rice (75).

Rhamnogalacturonan II, or RG-II, is another pectic polysaccharide originally found
in suspension cultured sycamore cells (19). It is structurally very different from RG-I.
RG-II is much smaller than RG-I (degree of polymerization ~ 60), and has a higher
proportion of rhamnosyl residues which are 3, 3,4, 2,3,4 and terminally linked instead
of 2 and 2,4 linked as in RG-I. In addition, RG-II has many unusual glycosyl residues
including 2-0-methylfucosyl, 2-0-methyl-xylosyl, apiosyl and 3-C—earboxyl-5-deoxy-L-
xylosyl (acetic acid) residues. RG-II was also found in the primary cell walls of other
dicots such as pea, pinto bean and tomato (19) which may suggest that RG-II or similar
pectic polysaccharides are common in dicot plant cell walls. .

Apiogalacturonans are pectic polysaccharides isolated from cell wall of Lanna minor
(8, 30, 31) and contain mainly galacturonosyl and apiosyl residues. The apiose content
in different fractions varied from 7.9% to 38.1% (30). In one ease in addition to the
apiosyl residues, which composed 25-28% of the apiogalacturonans, small amounts of
xylosyl and galactosyl residues were also reported to be present (8). Apiogalacturonans

are considered as ”unconventional pectins" (35), "apparently rare polymer” (21), and

15
'pectins of unusual composition” (36) beeause the apiosyl residue is one of their main

components and beeause the rhamnosyl residue a common residue of RG-I and RG-II,
has not been detected in them. Apiose, however, has been found to be a component of
polysaccharides in a number of different plants such as Posidonia australis (10), Elia sp.
(5), Zostera marina (5, 58, 81), Zostera nana (23), Zostera paciflca, Phyllospadix (58),
Lenora gibba (9) and Acer pseudoplatanus (suspension-cultured syeamore) (19). In
addition, Duff (23) reported that of 175 plant species examined, 31 showed ”traces"
amounts, 51 had "moderate” amounts and 17 were ”good sources” of apiose. Although
Duff did not distinguish between apiose in glycosides and that in polysaccharides, his
findings indieated that apiose must be relatively widely distributed in the plant kingdom.
However, the apiose content in polysaccharides could be underestimated or the sugar may
not even be detected by researchers who determine sugar composition by using GC-MS
analysis of alditol acetates, since apiitol acetate has a very similar retention time (on the
common GC columns used to separate alditol acetates) and mass spectrum to xylitol
acetate (19, 38). When apiogalactcuronans were partially characterized a number years
ago, paper chromatography was used to determine their glycosyl compositions (8, 30,
31). In one ease they were characterized as unesterified a-l,4 linked polygalacturonic.
acids to which monomeric side groups (75 % apiose and 25 % xylose residues) were
attached (8). In another ease both D—apiose and apiobiose (31) are attached to the
galacturonan with only 1.0-3.5 % of the galacturonosyl residues esterified. Rharnnose was
not reported to be present. Pectic polysaccharides of high apiose content are resistent t0
pectinaso-catalyzed cleavage (30, 31). Since a lack of rhamnose in these polysaccharides

would make them exceptions to the more common pectic polysaccharides and since a

16
number of techniques for characterizing polysaccharides have been improved or

developed since then, a re-examination of their structure was undertaken.

Physical and Chemieal Properties of Pectic Polysaccharides

Pectic polysaccharides are polyanionic molecules at most pH’s. When their main
chains interact noncovalently to form a three dimensional network that entraps solvent
and solute molecules, gels are formed. The property of the gel is dependent upon the
formation of " j unction zones” . These are zones where two or more chains or segments
of chains are in association through intermolecular hydrogen or ionic bonding (62-64, 66,
67). The greater the number and length of the junction zones, the more rigid is the gel.
If too many junction zones exist, precipitation rather than gel formation occurs. The size
and number of the junction zone depends on the structure of the pectic polysaccharide,
its concentration, types of cations present and their concentration, pH of the solution and
temperature. Pectic polysaccharides are soluble in pure water. In aqueous solutions, the
polygalacturonic acid portion of the backbone folds into a 2,-helix (2 stands for two
galacturonosyl residues per conformational repeat unit; the subscript, 1 stands for one
turn of the helix per repeat) conformation which has a zig-zag type of arrangement
involving an oxygen bridge between neighboring galacturonosyl residues (53, 65) (Figure
5A). Pectic polysaccharides form two types of gels, the calcium gel and the acid gel (11,
35). For the Ca2+ gel, the level of bound Ca2+ decreases with increasing concentration
of competing monovalent ions, until it reaches a constant value of ~ 50% of the total
stoichiometric equivalent of carboxyl groups (28) (Figure 5 C, top left). Ca2+ in the

solution causes the formation of ”egg-box”-like dimers (65) (Figure 5B) among

17

Figure 5. Proposed Structures of the Polygalacturonate Portion of the Pectic
Polysaccharide Backbone. (A) Structure of the polygalacturonate portion of the pectic
polysaccharide backbone. From reference (53). (B) Structure of the "egg-box" dimer
formed by two polygalacturonate chains and associated Ca2+ ions. Left, schematic
representation of two buckled chains (each by a zig-zag line) with Ca2+ ions represented
by circles. Right, detailed conformational representation of a possible mode of eation
coordination by one chain. From reference (65). (C) Orientation and stoichiometry of

Ca2+ binding within ordered interchain junctions. From reference (53).

18

{I HO c00' H0 COO‘
f o
0 0 Ho 0 HO
HO 0 HO 0
, 0 _ HO 0 HO 0\
COO c00‘

 

 

 

 

 

  

Lorne We

-> i007. Co" binding
0 = ea"

2, dimer
50% Cc" binding

906 ® a U6 has“ 336 63 91:3

3| dimer
~33°lc Co" binding

l9
unbranched, unesterified segments of the polygalacturonosyl chains, and the univalent

cation competitors such as Na+ prevent the dimers from aggregating further. The dimers
consist of two 21-helix chains or segments linked by calcium bridges with the carboxyl
groups on one side of one 21-helix sharing calcium ions with the carboxyl groups on the
opposite side of the other 21-helix through ionic bonds (53, 65). The earboxyl groups on
the dimer on the side away from the Ca2+ ionic bounds are free to associate with the
univalent eation competitors. If ealcium is the sole or principal eation in solution, the
dimers can aggregate further without rearrangement by using the free earboxyl groups
on the opposite side of each helix chain to form Ca2+ bridges with other dimers. This
would eventually lead to an approximate doubling of the amount of Ca“ bound
cooperatively and to the formation of sheetlike aggregates (53) (Figure 5C, top right).
If the aggregates are dried to a solid, the conformation of the polygalacturonosyl chains
changes to a 31-he1ix due to a polymorphic phase transition (53, 77) (Figure 5C, bottom
right). Side chains of neutral glycosyl residues affect the formation of the junction zones
(1 1). Rhamnosyl residues found in the backbone of pectic polysaccharides interrupt
junction zones through their alteration of conformation i.e. their 1,2-links form kinks in
the chains (35, 60). Many of the rhamnosyl residues have a variety of side chains
attached to them and these are considered to interfere with formation of junction zones,
although they may also interact noncovalently with each other to enhance gelling (4, 35).
The formation of a Ca2+ bridge junction zone requires a sequence of at least seven
consecutive free carboxyl groups along the participating face of each chain (60). The
esterification of the carboxyl groups interferes with formation of Ca2+ junction zones

(Fig. 6) and makes the gel network more flexible. At constant pH, the gel strength of a

20

Figure 6. Schematic Illustration of the Structure of the Ca“ Gel Network. From

reference (67).

 

21

 

  

or by highly esteritied
chain sequences

 

 

Site-bound calcium ions

 

 

 

22
Ca2+ gel increases as the content of methyl esterified groups decreases (67), although it

has been reported that some degree of methyl esterifieation ean be accommodated within
the junction zones of such gels if one of the two chains in the junction zone is not
esterified (60). The acetyl substituents on the hydroxyl groups of galacturonosyl residues
can also affect the binding of Ca“ by the carboxyl group presumably through steric
hinderance (41).

Pectic polysaccharides ean form another type of gel, the "acid gel” . This gel is
formed at low pH without the participation of Ca2+ . Pectins are highly ionized in solution
at neutral and alkaline pH, and the charges along the molecules keep them free from
aggregation by columbic repulsion. At a sufficiently low pH, the number of charges is
reduced enough so that sufficiart interchain hydrogen bonds are formed between two
hydroxyl groups, between a hydroxyl and an uncharged carboxyl group or between two
uncharged earboxyl groups or some combination to produce junction zones and an acid
gel (1 l). Pectins with higher degrees of methylesterification have less charge so they can
form an acid gel at a higher pH. At constant pH, the gel strength of an acid gel increases
with increasing DB of the pectin. The apparent pK of the carboxyl group of pectin with
a DB of 65 % is 3.55 while that of pectic polysaccharides with no methyl esterification
is 4.10 (59). Increasing the concentration of a monovalent cation ,such as Na+ and K“,
will result in decreasing hydrogen bonding in junction zones thereby decreasing gel
strength and reducing precipitation, and as a result a "salting in” occurs, although too
high a concentration of the monovalent cations may cause precipitation (32).

In solution, pectins are most stable at a pH of about 4. Both higher and lower pH can

cause deesterifieation and depolymerization (including hydrolysis) (l , 2, ll, 12, 40, 78),

23
however the present of solutes (i.e. cations) lowers water activity so that the rates of both

reactions are reduced (1 1). At pH values of 5~6, pectins are stable only at room
temperature. As the temperature is raised, pectin chains with methyl esters ean be
cleaved by B-elimination (2, 40). Some of the glycoside bonds in pectic polysaccharides
are also quite labile although the glycoside bond of the uronosyl residues is particularly
stable to acid hydrolysis. For example, hydrolysis of the side chains of apiogalacturonans
is substantially completed in 3 h at pH 4.5 and 100°C (31).

Since the stability and solubility of pectic polysaccharides in solution change with
changes in concentration, pH, and concentration and type of cation, great care must be
taken to avoid degradation, precipitation and gel-formation through the entire isolation,

purification and characterization process.

PART 1
ISOLATION AND HOMOGENEITY OF THE PECTIC POLYSACCHARIDES

FROM L. minor

MATERIALS AND METHODS

Materials: L. minor (duckweed) was grown as described elsewhere (39). DEAE
Sephadex A-25 and Dextrans T- 10, T40, T-70 and T-500 were obtained from Pharmacia
Inc. Sephacryl S-400 was obtained from Sigma Chemieal Co. 3-Pheny1phenol (3-

hydroxydiphenyl) was obtained from Aldrich Chemical Company.

General Methods: DEAE-Sephadex A-25 was used for all ion exchange chromatography.
Column fractions were assayed for uronic acid or total carbohydrates or both by the
method of Blumenkrantz and Asboe-Hansen (l4) and Dubois et al. (22), respectively.
Appropriate column fractions were combined, and dialyzed against double distilled water
for about 36 h with seven water changes. Dialyzed samples were concentrated in vacuo

at less than 35°C.

Isolation of Pectic Polysaccharides: Cell walls were prepared from fresh L. minor
(164.2 g, wet weight) following the procedure of Kindel (unpublished). Basieally the
procedure involved homogenizing the plants with a Waring Blendor in 1.0 M NaCl at
4°C, centrifuging, washing the plants with 1.0 M NaCl, centrifuging, and washing a

final time with water. The residue was called washed cell walls, and was extracted with

24

25
0.05 M ammonium oxalate (pH 6.5) at 22°C for 15 min and centrifuged. The

supernatant solution was collected, dialyzed, and the dialyzed solution was concentrated
and ealled sample A- 15m. The residue was resuspended in 0.05 M ammonium oxalate,
stirred at 22°C for 4.75h, centrifuged, and the supernatant solution was treated as above
and called A-4.75h. A complete second isolation was performed and the samples were
called B-15m and B-4.75h.

Column chromatography of the above four samples was performed on DEAE-
Sephadex with 0.067M sodium phosphate (pH 7.7) as the column buffer as described by
Hart and Kindel (30) except for the following changes: For the A- 15m and B-15 m
samples, the column was 1.9 cm (i.d.) x 14 cm (h) and the flow rate was 0.43 mllmin.
For the A—4.75h and B-4.75h samples the column was 1.7 cm (i.d.) x 14 cm (11) and the
flow rate was 0.3 mllmin. Before applieation to a column, each sample was filtered
through a membrane filter with a 1.2 u pore size. Samples and column fractions were
assayed for uronic acid. Recovery of material from the columns was based on the uronic
acid test. The following major samples were obtained: sample A-15 m yielded samples
A-lSm-II and A-lSm-III, sample A-4.75h gave sample A-4.75h-II, sample B-15m gave
samples B-lSm-II and B-lSm-III, and sample B-4.75h gave sample B-4.75h-II.

Samples from the first column were rechromatographed individually on a second
column of DEAE-Sephadex. The columns (1.35 cm, i.d. x ~ 14.2 cm, h) were
developed first with above column buffer and then with a linear gradient of NaCl in
column buffer. For samples A-lSm—II, A-4.75h-II, B-lSm-II and B-4.75h-II the NaCl
gradients were 0.075 M to 0.35 M; for samples A-lSm-III and B-lSm-III, the NaCl

gradients were 0.1 M to 0.4 M. Column fiactions (10 ml) were collected and analyzed

26
for uronic acid and total sugar. Fractions were combined and dialyzed. Polysaccharides

obtained from this re-chromatography were examined for homogeneity by

chromatographing each on DEAE-Sephadex and Sephacryl S-400.

Homogeneity of Polysaccharides: Chromatography of the above samples on DEAE-
Sephadex with 0.067 M sodium phosphate pH 7.7 as the column buffer, was performed
as described above except for the following changes: Each column was 0.95 cm (i.d.)
x ~12 cm (h); fractions of ~2.5 ml were collected at a flow rate of ~0.1 mllmin, and
the linear gradient of NaCl varied from starting concentrations of 0.05 to 0.1 M to final
concentrations of 0.3 to 0.4 M, depending on the individual sample (see Results).
Polysaccharide samples were also examined by column chromatography on' Sephacryl
S400 (1.0 cm, i.d. x 61.2 cm, h). Polysaccharide samples were concentrated to
between 0.2 and 0.4 ml (the range of the polysaccharide concentrations in the
concentrated samples based on the uronic acid test was 0.3 to 1.6 mg/ ml) before being
applied to the column. The column was developed with 0.01 M sodium phosphate (pH
6.5) at a flow rate of approximately 0.1 mllmin. Fractions of 1.5 ml were collected, and
the appropriate fractions were combined, dialyzed and stored at -20°C. Blue dextran and
K,Fe(CN), were used to obtain the void (v0) and total volumes (v.). Samples (0.2 ml,
2.5 mg/ml) of Dextran T-10, T40 and T-70 and T-500 (3.75 mg/ml) were
chromatographed individually on the column and the column 'fiactions were tested for

total sugar.

RESULTS

Chromatography on DEAE-Sephadex — Step Gradient: The column Chromatograms of
samples A-lSm, B-lSm, A—4.75h and B-4.75h are shown in Figure 7, A - D,
respectively. For the 15m samples, the majority of the material eluted with 0.2 and 0.25
M NaCl (Figure 7, A and B). Sample A-lSm yielded two major fractions, A-lSm-II and
A-lSm-III, that contained 28.2% and 41.8%, respectively, of the material recovered.
Sample B-15m also yielded two major fractions, B-lSm-II and B-lSm-III, that contained
26.8% and 50.5%, respectively, of the material recovered. For the 4.75h samples, most
of the material was eluted with 0.2 M NaCl. Fractions A-4.75h-II and B—4.75h-II
contained 72.8% and 75.9%, respectively, of the material recovered. The average
recovery of material from the columns (based on the uronic acid test) was 57.8% (range

was 30.3% to 93.2%).

Second Chromatography on DEAE—Sephadex — Linear Gradient: The column
Chromatograms of samples A-lSm-II, A-lSm-III, A-4.75h-II, B-lSm-II, B-15m-III and
B-4.75h—II on DEAE-Sephadex are shown in Figure 8, A - F, respectively. The average
recovery of sample from the columns was 92.1%. Sample A-lSm-II (Figure 8A) and B-
lSm—II (Figure 8D) both yielded two incompletely separated fractions on
rechromatography. They were treated separately and combined as A-15 m-IIa and A-15m-
IIb, and B-lSm-IIa and B-lSm-IIb. Rechromatography of sample A- 15 m-III showed a
single peak with a visible shoulder (Figure 8B). Fractions on the shoulder were opaque

and had a slight precipitate, and were combined separately as sample A-15 m-IIIb. Sample

27

28

Figure 7. Column Chromatograms of Cell Wall Pectic Polysaccharide Samples.
Chromatography was on DEAE-Sephadex as described in the Materials and Methods.
Panels A, B, C and D are Chromatograms of samples A-lSm, B-lSm, A-4.75h, and B-
4.75h, respectively. Fraction were tested for uronic acid (a ). The samples obtained are

named as indicated on the graphs.

Absorbanca at 520 nm

Absorbanee at 520 nm

 

2.0

 

 

 

     

‘ Builetl Mum lama-oil noun-cl I“:
Nun-II
A

  

 

is- H “mm
1
1.0-
II
4
0.5-
I
t bun-I
0c . . . .
o 20 40

j
‘0 .0

Fractlon number (15 mlltrsctlon)

 

 

 

 

O 20 40

Fraction number (15

 

mlltractlon)

 

29

Absorbancs at 520 nm

Absorbance at 520 nm

 

 

 

 

 

Fraction number (15Itractlon)

 

 

M78541

Mfld‘ smart-oldest»

masons-ct

 

 

 

10 20

'l‘

'l‘

 

40 50

Fractlon number (15 mlltractlon)

 

30

Figure 8. Column Chromatograms of Rechromatograpned Pectic Polysaccharide
Samples. Rechromatography was on DEAE-Sephadex as described in the Materials and
Methods. Panels A, B, C, D, E, and F are Chromatograms of samples A-lSm-II, A-lSm-
III, A-4.75h-II, B-lSm—II, B—lSm-III, and B-4.75h-II, respectively. Fractions were tested
for uronic acid (a ) and conductivity (0 ). Fractions between arrows (e.g. §<—'—>§ )

were combined as individual samples.

Absorbanco at 620 nm Absorbance at 520 nm

Abaorbanco at 520 nm

 

 

 

 

      

  
 

 

 

    

0.3 0.4
«VI
”0.3
02‘
1 ”0.2
p
Q1-
. '0.1
o
0
d o
I
. F
' b
' a
M" v 1 V 1i V A ”0.0
0 10 20 30 40

Fraction numbers (10 mlliractlon)

 

1.2

1

 

 

O 10 20

 

 

 

80 4O

Fraction numbers (10 mlliractlon)

6.3

 

 

 

 

o r o 20
Fraction numbers

”0.2

 

 

 

30 40
(2.5 ml/iractlon)

31

NsCl (M)

Neel (M)

".01 (fl)

Absorb-nee at 620 nm Absorbancs at 520 nm

Absorbancs at 520 nm

 

 

 

 

 

 

 

10 20 30

40
Fraction numbers (10 mllirsctlon)

0.4

 

 

 

1

 

8015014"
0 " v - 1 -

1 0 20 30

 

40

Fraction numbers (10 mlliractlon)

Q3

 

 

  

 

  

"0.!

~04

 

 

10 20
Fractlo numbers (10 mllirsetlon)

Meet (it)

met (in)

NICI (fl)

32
A-4.75h-II was separated into three fractions: A—4.75h-IIa, A-4.75h-IIb and A-4.75h-IIc

(Figure 8C). Sample B-lSm-III showed a major peak with a incompletely separated
minor peak (Figure 8E). Because there was a drop in the rate of increase of the NaCl
gradient followed by a relatively sharp increase as the minor peak was eluting, the small
peak was considered an artifact and was combined with the larger peak as sample B-15m-
III. Rechromatography results presented later showed sample B-lSm-III eluted as a single
band from a DEAE-Sephadex column. Five incompletely separated peaks were found on
rechromatography of sample B-4.75h-II (Fig. 8 F). They were divided into samples B-

4.75h-IIa and B-4.75h-IIb.

Homogeneity of polysaccharides: DEAE Sephadex column chromatograms of samples
A-lSm-IIa, A-lSm-IIb, A-4.75h-IIa, A-4.75h-IIb and A-4.75h-IIc are shown in Figure
9, A - B, respectively, and those of samples B-lSm-IIa, B-lSm-IIb, B-lSm-III, B-4,75h-
11a and B-4.75h-IIb are in Figure 10, A - B, respectively. Column chromatograms of
samples A-lSm-IIa, A-lSm-Ilb, B-lSm-IIa, B-lSm-Ilb, B—lSm-III, B-lSm-4.75h-IIa and
B-4.75h-IIb were also obtained by testing for total sugar and a comparison of these
results and those obtained with the uronic acid test (Figure 9, A and B and Figure 10,
A - E) is shown in Figure 11, A, B, E, F, G, H, and 1, respectively. The results showed
that each sample eluted from DEAE—Sephadex as a single, major peak and that the total
sugar and uronic acid chromatograms were basically coincident. However, for samples
A-lSm-IIa and A-lSm—IIb, some small neutral sugar peaks were presented in early
fractions (Figure 11, A and B). Samples A-lSm-IIIa and b were not chromatographed

a third time on DEAE-Sephadex because when together as sample A-15 m-III (Figure

33

Figure 9. Column Chromatograms of Pectic Polysaccharides Samples Obtained
from Sample A and Chromatographed Twice Previously on DEAE-Sephadex.
Chromatography was on DEAE-Sephadex as described in the Materials and Methods.
Panels A, B, C, D, and E are chromatograms of samples A—lSm—IIa, A-lSm-IIb, A-
4.75h-IIa, A—4.75h-IIb, and A-4.75h—IIc respectively. Fractions were tested for uronic

acid (I) and conductivity (0 ).

Absorbance at. 620 nm

Aboorbanoe at 520 nm

 

0.6

0.5“

0.4 1

 

Fraction Number (2.5 ml/fraction)

10 20

30

 

40

 

0.3

 

1.2

4

1.0"

 

0.0“
0

 

 

10 20

30

 

*0.0
40

Fraction Number (2.5 mllfraction)

Absorbance at 520 nm

34

.NaCl (M)

NaCl (M)

Absorbance at 520 nm

Absorbance at 520 nm

 

0.8

0.6 7

0.4 '

 

0.0 "
0

40

 

 

Fraction Numbers (2.5 ml/fraction)

 

2.5

J

 

0.0 ‘
0

10

 

 

20

30

i
~02

r0.1

 

'0.0
40

Fraction Number (2.5 mllfraction)

0.3

 

2.0

1.5“

1.0"

0.5“

 

0.0 ‘
0

10

 

 

20

30

i

" 0.2

 

I’0.0
40

Fraction Number (2.5 mllfraction)

NaCl (M)

NaCl (M)

NaCl (M)

35

Figure 10. Column Chromatograms of Pectic Polysaccharide Samples Obtained
from Sample B and Chromatographed Twice Previously on DEAE-Sephadex.
Chromatography was on DEAE-Sephadex as described in the Materials and Methods.
Panels A, B, C, D, and E are chromatograms of samples B-lSm-IIa, B-lSm-Ilb, B-15m-
III, B-4.75h-IIa, and B—4.75h-Ilb respectively. Fractions were tested for uronic acid

( ) and conductivity (0 ).

 

Absorbance at 520 nm

Absorbanee at 520 nm

 

 

0.0 7
0

 

10

 

20 30

Fraction Number (2.5 mllfraction)

 

 

 

 

 

 

 

 

3.0 0.4
2.5 “
" 0.3

2.0 "‘
1.5 “ "' 0.2
1.0 "‘

‘ " 0.1
0.5 "

t i
0.0 " T " 0.0

0 1 0 20 30

Fraction Number (2.5 nil/fraction)

Absorbance at 520 nm

NaCl (M)

NaCl (M)

36

 

 

 

 

 

 

 

 

2.0 0.4
B
E 1.5" "0.3
0
ID
1'3
1.0“ "0.2
8
5 i
'5
g 0.5‘ "0.1
i I [
0.0' 1'0.0
0 1 O 20 30
Fraction Number (2.5 ml/fraction)
2.0 0.4
D I
E 1.5- Los
c
or . .
IO
‘3
a
1.0' "0.2
8
i , .
o
5 0.5- -o.1
. r
0.0' '0.0
0 1 0 20 30

I'h'action Number (2.6 mllfraction)

 

 

 

0.0"
0 10

 

NaCl (M)

 

20 30

Fraction Number (2.5 mllfraction)

NaCl (M)

NaCl (M)

37

Figure 11. Comparison of Column Chromatograms Obtained by the Uronic Acid
and Total Sugar Tests. Panels A, B, C, D, E, F, G, H, and I are chromatograms of
samples A-lSm-IIa (Figure 9A), A-lSm-Ilb (Figure 9B), A-lSm-III (Figure BB), A-
4.75h-II (Figure 8C), B—lSm—II-a (Figure 10A), B-lSm-Ilb (Figure 10B), B-lSm-III
(Figure 10C), B-4.75h-lIa (Figure 10D), and B-4.75-Ilb (Figure 10E) respectively.

Fraction were tested for uronic acid (a) and total sugar (A ).

38

Geno-g:— 93 3:5: ornat—

 

 

 

 

 

o... 8 3 o
unri « ii: i . .o
L

_

7
is

v
.a

H v
e

2.3% 9% 84:2. 5qu
8 on o. o
...... . - 13

   

 

to;

 

 

sewage o: 895:. 898.5
3 8 8 2

 

 

 

 

 

 

mm- m

moi.

3998::— 05 35.:- Bin
on on o p o

 

 

 

238% 3 sin: 38;
na cm or o- a o

 

 

 

 

333.58 3 36.... 5:8:
3 o. 8 2 o

   

 

 

 

NJ. mom m

M‘J.

 

Ecru-+58 95 362. .33
3 2 3 ..

 

 

 

 

“338% 3 93:5: «8399‘

ov ca on o— 00

NJ.

 

 

 

 

€58.58 ad Sea: 8:25
3 on 8 o F o

     

m

 

 

39
8C), they basically chromatographed as a single band and the chromatograms from both

uronic acid and total sugar tests were coincident (Figure 11C). In further work they were
treated separately. Samples A-4.75h-IIa, A-4.75h-IIb and A-4.75-I1c (Figure 9, C - B)
were also not tested for total sugar since sample A-4.75h-II (Figure 8C), the sample from
which they were derived, had already been tested for both uronic acid and total sugar and
the chromatograms were coincident (Figure 11D).

Purified samples were chromatographed on Sephacryl S-400, and each showed a
single band (Figure 12, A - L) except samples A-lSm-Ilb (Figure 12B), A-lSm—Ila
(Figure 12C) and A-15 m-IIIb (Figure 12D) which showed either a small peak just before
or after the major peak. For samples A-lSm-IIIa and A-lSm—IIIb (Figure 12, C and D),
it is possible that the small peaks are due to cross contamination of each by the other
(Figure 13). Sample A-lSm-IIlb was a shoulder in the parent peak (Figure BB)

K" values were calculated for all samples and standard dextrans as follows:
K _ VC- V0
av Vt- Va

 

where KIv = the partition coefficient of the sample; V, = void volume;

V, = elution volume of the sample; V, = total volume of the gel bed
K" values of the standard dextran samples were plotted against the logw of their weight
average molecular weight, M. (Figure 14). The K" value of each pectic polysaccharide,
the weight average molecular weight that a standard dextran with the same Km, would
have, and the NaCl concentration at the mid-point at which each pectic polysaccharide

' peak eluted from DEAE- Sephadex columns 2 and 3, are given in Table 1.

40

Figure 12. Column Chromatograms 01' Purified Pectic Polysaccharide Samples
Obtained with Sephacryl S-400. Chromatography was performed as described in the
Materials and Methods. Panels A, B, C, D, E, F, G, H, I, J, K, and L are
chromatograms of samples A—lSm-Ha, A-15 m-Ilb, A-lSm-IlIa, A-lSm-IIIb, A-4.75h-IIa,
A-4.75h-IIb, A-4.7Sh-IIc, B-lSm-IIa, B-lSm-Ilb, B-lSm-III, B-4.75h-IIa, and B-4.75h-

IIb, respectively. Fraction were tested for uronic acid (a ).

Abeorbence at 520 nm Abeorbance at 520 nm

Abeorbence at 520 nm

Abeorbanoe at 520 nm

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

15
i
i A
10‘
0.5‘
J
wo lb ab oh no
”action Number (1.5 nil/fraction)
05
05‘
I)
Q4'
034
J
021
01*
_- A.
' o lo 20 so co
Men Number (1.8 mvrraction)
3
(}
2 4
‘ q
0 . v -
0 10 20 30 40
Fraction Number (1.5 mlll'raen'on)
Q:
J'
01‘
0.2‘
03 -‘- v . v —
0 10 2O 30 40

hection Number (1.5 nil/fraction)

Abeorbanoe at 520 nm

Abeorbance at 520 nm

Abeorbanee at 520 nm

Aboorbance at 520 nm

41

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

15
I3
10‘
05‘
1
06 . . - u
0 10 20 30 (0
Fraction Number (1.6 mint-action)
15
13
1.0-4
0.51
J
I
__ ‘n‘ A “L
' o 10 to so so
Fraction Number (1.5 midi-action)
10
1
an H
05‘
J
I
01‘
J
02‘
1
Q6 v -w 4+ -
0 10 20 30 40
Fraction Number (1.5 mllfraction)
25
20‘ I(
4
154
J
10‘
05‘
J
00 v —~7‘ v ‘-
0 10 20 30 40

Fraction Number (1.6 ml/iraetion)

Abeorbance at 520 nm

Abeorbence at. 520 nm

Abeorbence at. 620 nm

Abeorbanee at 520 nm

 

A

 

 

10 F; 30
Mon Number (1.5 mlll'raction)

"-"w'

 

so

 

 

 

J

 

 

0 - v .
10 20 IO
Fraction Number (1.5 mVi’ractien)

40

 

20

 

 

 

13 53 5b
Fraction Number (1.5 mllii‘action)

40

 

30

25'

20‘

 

 

 

13 ab 5;—
Fraction Number (1.5 mllfraction)

40

42

Figure 13. Comparison of Column Chromatograms of Samples A-lSm-IIIa and A-
15m-IIlb. The column chromatograms of samples A-lSm-IIIa ( a ) and A-lSm-IIIb
( o ) obtained with Sephacryl S-400. Fractions were analyzed for uronic acid as

described in the Materials and Methods.

 

Absorbance at 520 nm

43

 

06-

Q4-

02-

 

0.0 '1

f“ _
u

 

 

v

,.
'- , .u L
A ' d
J l
/ —~ :
’ ‘ —l_'__. —a
‘ j ' .1 u u u U

I

10 20 30 40

J

Fraction number (1.5 mllfraction)

44

Figure 14. Relationship between M, of Standard Dextrans and K... The equation

for the line is: y = 1.71 - 0.254 x, where y = K,v and x = logme

ie (I:

Kav

45

 

0.8

0.7 "

0.6 "

0.5 "

0.4 "

0.3 "

 

‘ (— Dextran T-lO

    
    

‘— Dextran T-40

Dextran T-70 -——)

Dextran T-500 ———)

 

0.2

 

46
Table 1. Comparison of NaCl concentrations needed to elute polysaccharide samples

from DEAE—Sephadex, relative mobilities, as K", of samples on Sephacryl S-400, and

the expected weight average molecular weight of a standard dextlan with the same K...

   
  

 

 

 

 

 

 

 

 

 

 

 

 

 

DEAE Sephadex
chromatography, NaCl Sephacryl S-400 chromatography
Pectic polysaccharide molarity at ‘P‘d'win‘
sample of elutlon
Column 2 Column 3 K“ Expected Mdof standard

dextran with this X,
A-lSm-Ha 0.160 0.189 0.650 14,900
A-lSm-Ilb 0.282 0.226 0.527 45,500
A—lSm-Illa 0.242 Not deter. 0.361 205,000
A-lSm-lIIb 0.259 Not deter. 0.404 139,000
A-4.75h-Ila 0.083 0.051 0.673 12,100
A-4.75h-Ilb 0.112 0.119 0.711 8,650
A-4.75h-Ilc 0.145 0.150 0.656 14,200
B-lSm-lla 0.168 0.187 0.553 15,500
B-lSm-IIb 0.209 0.231 0.500 26,000
B-lSm-III 0.220 0.240 0.316 159,000
B-4.75h-Ila 0.117 0.136 0.647 6,120
B-4.75h-IIb 0.155 0.168 _ 0.579 12,000

 

 

 

 

 

 

DISCUSSION

All cell wall pectic polysaccharide samples of L. minor isolated proved to be
homogenous in terms of charge content, since each eluted as a single, reasonably narrow
and symmetrieal band from a DEAE-Sephadex column, and the chromatogram profiles
from the uronic and total sugar tests were largely coincident. There were no significant
neutral sugar peaks detected. The small peak of neutral sugar found in the early column
fractions of samples A-lSm-IIa and A-lSm-IIb (Fig. 9, A and B) may indicate a loss of
side chains. In addition, all the samples analyzed by gel filtration column
chromatography with Sephacryl S-400 also showed a single, reasonably symmetrieal
pectic polysaccharide band except for samples A-15 m—IIIa and A- 15 m- IIIb, which
probably are cross contaminated with the each other, and sample A-15 m-IIb, which has
a small shoulder. This indicates they all are homogenous or reasonable so with respect
to size. However, each sample very likely consists of a population of molecules in which
there is a range of molecular sizes that has a Gaussian destribution. However, it is within
this context that they are considered to be homogenous.

Generally similar results were obtained from ion exchange and gel filtration
chromatography of cell wall pectic polysaccharides from both isolations. For samples A-
lSm—IIa, A-lSm-IIb, A-4.75h-IIb and A-4.75h-IIc, the molality of NaCl at the mid-point
of elution from the DEAE-Sephadex columns and the K" value obtained from
chromatography on Sephacryl S-400 are both similar to the values for samples B-15m-
IIa, B-lSm-Ilb, B—4.75h-IIa and B-4.75h-IIb, respectively (Table 1), although one more

small sample peak, A—4.75h-IIa, was separated from sample A-lSm—II (at low NaCl

47

48
concetration, 0.083 M) than from B-15 m-II. The NaCl molarities for elution and the K“,

of samples A-lSm-IIIa and A-lSm-IIIb were also similar to that of sample B-lSm—III.
Comparison of the pectic polysaccharide samples extracted in 15 min with those
extracted in the next 4.75 h showed the latter generally have a smaller molecular size and
eluted from a DEAE-Sephadex column at a lower NaCl concentration. A examination of
the purified samples isolated from a single cell wall preparation shows that the larger the
molecular size (smaller K" value) of the purified sample, the higher the concentration
of NaCl needed to elute it from the column. The exception is sample A-lSm-Illb, whose

molecular size is smaller than A- lSm-IIIa but it eluted at a higher salt concentration.

PART 2

SUGAR COMPOSITION OF PECTIC POLYSACCHARIDES

MATERIALS AND METHODS

Materials: CMC, NaBD, (98 atom 96 D), NaBH., TFA and 3-phenylphenol (3-
hydroxydiphenyl) were obtained from Aldrich Chemical Company, Inc.; 1-
methylimidazole and polygalacturonic acid were obtained from Sigma Chemical
Company; D—xylose, L-arabinose, L-fucose, L-rhamnose, D-galactose, D-mannose, and
myo—inositol were obtained from Pfanstiehl Laboratories, Inc.; D-glucose was from the
U.S. Department of Commerce, National Bureau of Standards. D-Apiose was isolated
from L. minor by method of Neal and Kindel (56). Capillary columns SP-2380 and DB-

225 were from Supelco, Inc. and J & W Scientific, Inc., respectively.

Reduction of Pectic Polysaccharides: The carboxyl group of the galacturonosyl residues
of pectic polysaccharides was reduced basically as described by Taylor and Conrad (74).
Dialyzed sample solutions of A-lSm-IIIa, A—lSm-IIIb, B-lSm-IIa, B-lSm-IIb, B-lSm-III,
B-4.75h-IIa and B—4.75h-IIb, each containing 1.2 - 4.7 mg of sample, were evaporated
to about 10 - 15 ml. After the CMC was added to the sample solution, the pH was
maintained at 4.75 for about 2.5 h by addition of 0.05 N HCl with a Sargent-Welch
Recording pH-STAT. After esterification with CMC, 250 - 400 mg of NaBD4 in 6 - 10
ml water was added dropwise over 1.5 h, and the sample was maintained at pH 7 with

5 N HCl and the pH-STAT. [The solution was transferred to dialysis tubing that retained

49

50
molecules with a molecular weight of 6000-8000 or larger and dialyzed against water for

36 h. All samples were reduced twice except sample B-lSm-III which was reduced three
times.

Before and after reduction, a portion of the sample solution was used for
determination of uronic acid and total carbohydrates by the method of Blumenkrantz and
Asboe-Hansen (14) and Dubois et al. (22), respectively. The amount of uronic acid and

neutral sugar in the pectic polysaccharides was ealculated by using the following equation

 

 

 

(derivation in, Appendix):
"5'0
Amr- M21):(Amt-bz—bbe-bs-brb, D1
c _ .
" l _ m, -[m., x+m8(l -x)] m, x+m6(l -x)

mz-[nls x+m6'(l-x)]
where c,I = concentration of the neutral sugar in the sample, units depend on units used
in standard curves
A.” T = absorbance of sample (including neutral sugar and uronic acid residues) by
the total sugar test
Am T = absorbance of the sample by the uronic acid test
b2 and ln2 = intercept and slope of the standard curve for uronic acid residue in the
uronic acid test
b, and m3 = intercept and slope of the standard curve for uronic acid residue in the total
sugar test.
b, and in, = intercept and slope of the standard curve for apiosyl residue in the total
sugar test

b, and m, = intercept and slope of the standard curve for galactosyl residue in the total

5 1
sugar test

b7 and m, = intercept and slope of the standard curve for apiosyl residue in the uronic
acid test

b, and m. = intercept and slope of the standard curve for galactosyl residue in the uronic
acid test

D1 = -fold dilution of the sample for the total sugar test

D2 = -fold dilution of the sample for the uronic acid test

x = proportion of the total neutral sugar residues that is apiosyl residue

Amount of uronic acid in the sample was ealculated by using the following:

c
Amt-brbs-[nu cum, (1 -x)] 3' -b2
ce- 2 .D 2
"'2

where c. = concentration of the uronic acid residue in the sample

 

A software program (written with BASIC) was developed to perform the necessary
ealculations. Based on the results for the uronic acid and total sugar determinations, the

total recovery of material from each reduction was calculated.

Analysis of Reduced Pectic Polysaccharides: Reduced samples were evaporate to about
1 to 2 ml and blown to dryness with N2. Myo-inositol (0.15. mg) was added to each
sample and the sample was hydrolyzed with 2 M TFA at 110°C for 1.5 h. A time course
of hydrolysis of a typical reduced pectic polysaccharide sample was performed. The
sample was treated with 2 M TFA at 110°C for 0.5, 1, 2, 3 and 4 h and after
hydrolysis, each portion was blown to dryness with N2 at 22°C.

The procedure of Blakeney et a1. (13) as modified by Kindel and Liang (38) was

52
used to prepare alditol acetates, except after acetylation, the sample in 1.5 ml of

chloroform was washed five times with water. After the last extraction, the solution was
centrifuged and the organic phase was transferred to a 1 ml vial, brought to dryness with
N, and the sample dissolved in 0.2 ml of chloroform and used for GC. A standard
sample mixture, containing rhamnose, fucose, arabinose, xylose, mannose, galactose,
glucose, myo-inositol (each 150 pg) and apiose (153.5 pg), was acetylated by the
procedure described above and washed with water 1, 2 and 5 times, and gas
chromatography was used to determine the effect of washing. The sample extracted 5
times served as the standard sample to obtain the gas chromatography response factor for
each sugar. Myo-inositol served as the internal standard.

Gas chromatography of alditol acetates was performed with a Hewlett Packard gas
chromatograph, Model 5480A, equipped with a flame-ionization detector, a 5840A data
terminal and a splitter from J & W Scientific, Inc. Samples were chromatographed on
a DB-225 fused-silica capillary column (30 m x 0.25 mm i. d., film thickness 0.25 pm)
isothermally at 230°C, with helium as the earlier gas at a flow rate of ~0.8 mllmin.
N, was the makeup gas and had a flow rate of 45 mllmin. Samples of 0.5 - 1.5 pl were
injected and the split ratio was 1:45.

EI mass spectrometry were performed with a IEOL JMS-AX505H mass spectrometer
interfaced to a Hewlett Packard gas chromatograph, Model 5890 A, which was equipped
with a splitless injector and the same type of DB-225 capillary column as above. Helium
was the carrier gas and the flow rate was ~l mllmin. The ion source temperature was
200°C and the ionizing voltage was 70 eV. Alditol acetate samples were diluted with

chloroform, and about 0.5 p1 was injected with a on-column injection system.

5 3
Alditol acetates were identified by retention time and ion patterns. The proportion

of galactitol acetate derived from galactosyl residues and that from galacturonosyl

residues was calculated from the equation (derivation in Appendix),

R- 2x(Ix+2-Ixxe,2) x 100
Ixx(1-Fn,)+1x,2 98

 

where R = the ratio of galactitol acetate molecules that are labeled with deuterium to
total galactitol acetate molecules. When the ion pair 217/219 mlz was chosen to represent
galactitol acetate/galactitol acetate-D2, I, and I“; were the intensities of ion 217 mlz (the
nominal ion, x) and 219 m/z (ion x containing two natural l3C, a 11‘O or two deuterium
atoms from NaBD4), respectively. FM2 is the theoretical abundance ratio of natural
isotope contributed to the x+2 ion which has a value of 0.0169 (the probability of two
13C in nine carbon atoms is (9)2 x 0.006% or 0.00486, and of one "0 in six oxygen
atoms is 6 x 0.2% or 0.012).

When the ion pair 259/261 mlz was chosen to represent galactitol acetate/galactitol
acetate D2, L and I", were the intensities of ion 259 mlz (the nominal ion, x) and 261
mlz (ion x containing two natural 13C, a 1“O or two deuterium atoms from NaBD4),
respectively. PM is the theoretical abundance ratio of natural isotope contributed to the
x+2 ion which has a value of 0.0213 (the probability of two 13C in 11 carbon atoms is

(11)2 x 0.006% or 0.00726, and of one 1“0 in 7 oxygen atoms is 7 x 0.2% = 0.014).

RESULTS

The unreduced uronic acid remaining after each reduction and the total recovery of
material from the reductions are shown in Table 2. After reduction was performed once,
the average uronic content was approximately 13 % (w/w) of the reduced pectic
polysaccharide samples (range from 5.6% to 23.4 % , excluding B-lSm-IIa). A second
reduction decreased this to 8.6% . All samples were reduced twice, except sample B-lSm-
III, which was reduce three times. Although reduction of pectic polysaccharides a third
time resulted in a further decrease in the uronosyl residue content (e.g. the third
reduction of sample B-15 m—III brought the uronosyl residue content from 11.1% to
5.5 %), each reduction resulted in loss of sample material and therefore, because of
limited sample quantities, a third reduction was not performed. The average amount of
sample material left after two reductions was 44% of the starting material (Table 2).

The time course of TFA hydrolysis (Figure 15) showed that the yield of arabinose
reached maximum within a half hour of hydrolysis, the apiose yield was highest after 1
hour of hydrolysis, for galactose, rhamnose and glucose 2 hours of hydrolysis were
needed to obtain the maximum yield, and the yield of mannosewas highest after 3 hour.

Gas chromatograms of the standard alditol acetate sample in chloroform after being
washed with water 1, 2, and 5 times are compared in Figure 16, A - C, respectively.
The tailing after the solvent peak for the sample washed twice (Figure 16B) is
signifieantly less than for the sample washed once (Figure 16A). After 5 washings, the
interference from tailing is negligible (Figure 16C).

Rhamnose, fucose, arabinose, xylose, apiose, mannose, glucose, and galacturonic

54

55
Table 2. Reduction of cell wall pectic polysaccharides samples. The neutral sugar (%) and

uronic acid (%) are weight/weight percentages of each residue in the reduced sample. The total
recovery is also a weight/weight percentage. These were determined and calculated as described

in Materials and Methods. Only sample B-lSm-III was reduced three times.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Sample Alter first reduction After second reduction Alter third reduction Total recovery
of neutral
Neutral Uronic Neutral Uronic Neutral Uronic “'3“: “'9
sugar (96) acid (95) sugar (95) acid (it) sugar (96) acid (95) mm “'4 (7‘)
A-15m-ma 87.7 12.3 95.7 4.3 40.4
A—15m-mb 86.2 13.8 86.4 13.6 62.1
B-lSm-Ila 61.2“ 38.8"I 86.0 14.0 43.8
B-lSm-Ilb 91.1 8.9 94.2 5.8 27.3
B-lSm-III 84.9 15.1 88.9 A 11.1 94.5 5.5 73.3
843511-118 76.6 23.4 93.3 6.7 27.2
343511-111) 94.4 5 .6 95 .6 4.4 34.1
Average 86.8“ 13.2“ 91.4 8.6 44.0

 

 

 

 

 

* Abnormal color (it was yellowish-green instead of the normal pink color of uronic acid) was
observed in the uronic acid test. Sample had a high starting neutral sugar content (Table 4).
‘"' For the reason stated above, the data from the first reduction of sample B-15 m-lIa were not

used for the calculation of average percentage of neutral sugar and uronic acid.

56

Figure 15. Time Course of Hydrolysis of a Carboxyl-reduced Pectic Polysaccharide
Sample. Reduction and TFA hydrolysis of the cell wall pectic polysaccharide sample and
preparation of alditol acetates and GC were performed as described in the Materials and

Methods.

 

 

57

  

0
6

000

1111

 

P
%\

  

\ \ wwwm
\ \ to

 

 

 

 

117.111.11.11}?

 

Q3 2&5 .5 20;

mMm

 

58

Figure 16. Treatment of an Alditol Acetate Sample with Water. Panels A, B and C
are gas chromatograms of a standard mixture of alditol acetates prepared as described in
the Materials and Methods and washed with water 1, 2, and 5 times, respectively.
Samples (2 pl) were chromatographed on a SP-2380 fused-silica eapillary column
isothermally at 260°C. The split ratio was 1:50. Other conditions were as described in
the Materials and Methods. The tailing caused by contaminants in the samples is
indicated by the arrow. Peaks 1 through 8 represent the alditol acetates of rhamnose (1),
fucose (2), arabinose (3), xylose (4), apiose (5), mannose (6), galactose (7), and glucose

(8) while peak 9 is myo-inositol hexaacetate.

 

 

 

 

 

59

 

 

 

 

 

 

 

 

 

e vi.»
2

 

 

 

 

 

 

 

 

 

 

 

e S rah.

J

 

 

 

 

 

 

 

an a- - a-.- o-- _— --—--

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

60
acid were detected in all seven cell wall pectic polysaccharides samples (Table 3). Small

amounts (4.1 to 11.3 mole %) of galactose was found in samples B-lSm-IIa, B-lSm-IIb,
B-4.75h-IIa and B-4.75h-IIb but not in the others. Each alditol acetate was identified by
matching its retention time to a known in the standard sample chromatographed under
the same conditions, and by comparison of its mass spectrum with that of the known.
The amount of each sugar in the sample was determined by the internal standard method
and the molar percentage of each sugar in the cell wall pectic polysaccharide samples is
shown in Table 3. Since samples A-15 m-IIIa and A-lSm—IIIb were obtained by dividing
sample A- 15 m-III during the isolation procedure and were analyzed separately , a
weighted average sugar composition was calculated to represent sample A-15 m-III (Table
3). The A-lSm-HI (ave.) and B-lSm-III samples are similar except that the apiose content
of the B sample is considerably higher than that of the A (ave.) sample.

A comparison of selected properties of the pectic polysaccharides samples to the
rhamnose/galacturonic acid ratio of these samples is shown in Table 4.

The ion clusters around the two ion pairs, 217/219 mlz and 259/261 mlz, were
analyzed to find the ratio, galactitol acetate/galactitol acetate-D2, which represents the
ratio of galactosyl/galacturonosyl residues in the pectic polysaccharide sammes. The
results (Table 5) showed that samples A-lSm-IIIa, A-lSm-Illb and B-lSm-III had 100%
of galactitol acetate as galactitol acetate-D2 (values greater than 100% were counted as
100% , since this is maximum possible), while samples B—lSm—Ila and B-lSm—lIb had low
values (galactitol acetate-D2 was 49 and 69 %). The percent galactitol acetate-D2 values

of samples extracted in the 4.75 h period were intermediate.

61
Table 3. Sugar composition of the cell wall pectic polysaccharide samples. The values

in the table are mole percents. The weight of each sugar detected by GC (internal
standard method) was divided by the molecular weight of the sugar to calculate the mole
percent. The mole percent of galacturonic acid (GalUA) and galactose were calculated
from the mole percent of galactose obtained from GC and the R value (shown in Table
4) of the sample. The amount of uronic acid remaining after reduction of the carboxyl
groups (calculated from the colometric tests) was assumed to be all galacturonosyl
residue and was used to adjust the mole percent of galacturonic acid.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Average"I
Sugar A-lSm A-15m of B-lSm B-15m B-lSm B-4.75h
-IIIa -lIIb A-15 m-IIIa -IIa -Ilb -111 -11a
A-lel-Illb
Rhamnose 4.6 2.8 3.8 3.0 4.5 3.0 4.0
Fucoee 1.3 0.80 1.1 1.5 1.5 1.5 2.2
r Arabinoee 2.5 1.5 2.1 4.0 3.5 1.4 2.8
Xylose 3.3 2.7 3.0 14.5 6.8 0.68 3.6
Apiose 32.5 16.3 25.6 23.7 23.0 39.8 39.0
Mannose 2.1 1.7 1.9 11.5 6.0 0.68 2.3
GalUA 44.8 67.7 54.6 20.8 30.2 51.3 38.8
Galactose 0 0 0 9.0 11.3 0 4. l
Glucose 9.0 6.5 7.9 1 1.9 13.4 1.6 3.2

 

 

 

 

 

 

 

 

* The average mole percent values for combined sample A-15 m-IIIa and A-lSm-IIb was

calculated as follows: _ Ma ma+Mbmb
M - x

ma+mb

100

 

where M = average mole percent of individual sugar residues in combined sample A-
15m—IIIa and A-lSm-IIb

M, and M, = mole percent of individual sugar residues in samples A-15 m-IIIa and A-
15 m-IIIb, respectively.

wt. and wt, = amount of samples A-lSm-IIIa and A—lSm-IIIb, respectively isolated from
the cell wall (see Figure 8B).

62
Table 4. Comparison of selected properties of the seven pectic polysaccharide samples

with the ratio of rhamnose/galacturonic acid (Rha/GalUA, mole! mole) for these samples.

 

 

 

 

 

 

 

 

 

 

I Time of Elution Relative

Sample Extraction Concentration Molecular
of NaCl (M) Size“

| A-lSm-llIa 15 min 0.25 205,000
A-lSm-III-b 15 min 0.25 139,000

I Average of A-lSm-IIIa & A-lSm-lllb" 15 min 0.25 177,000
B—lSm-III 15 min 0.25 159,000
B-lSm-Ila 15 min 0.20 15,500
B-lSm-Ilb 15 min 0.20 26,000
B-4.75h-lIa 4.75 h 0.20 6,000
B-4.75h-Ilb 4.75 h 0.20 12,000

 

 

 

 

 

 

* The relative molecular size of each sample is represented by the M. of a standard
dextran whose mobility is the same as that of the sample on a Sephacryl S-400 column.
** The average relative molecular size and ratio of Rha/GalUA of samples A-lSm-IlIa

and A-15 m-IIlb is weighted average (see Table 3).

63

Table 5. Intensity of ions in two major ion clusters from the mass spectrums of galactitol

acetate/galactitol acetate-D2 prepared from the pectic polysaccharides samples. The ratio

of galactitol acetate-Dzlgalactitol acetatea + galactitol acetate—D, (R) was calculated as

described in the Materials and Methods.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

I I
Sample A-15m- A-15m- B-lSm— B-15m- B-15m- B-4.75h- B-4.75h-
lIIa IlIh 11a 11b III 118 11b
217 10.3609 13.3649 16.2231 56.0119 56.9824 24.5813 53.3447
Ion mlz
Intensity
219 10.3311 14.3796 5.3405 28.3447 53.6743 18.4599 37.023
mlz
R 101.4 105.0 48.6 67.0 98.1 86.4 82.4
259 8.7105 10.4002 13.8702 40.5700 42.8466 15.3357 39.3615
Ion mlz
Intensity
261 8.2411 11.9290 4.7439 22.0947 44.7692 12.9675 27.6367
mlz
R 98.1 108.1 49.6 70.1 103.2 92.2 82.7 I
I Average R 100 107 49 69 101 éL89 83 I

 

DISCUSSION

Rhamnose, fucose, arabinose, apiose, xylose, mannose, glucose and galacturonic acid
were found in the all pectic polysaccharide samples isolated from the cell wall of L.
minor. Galacturonic acid (20.8 to 67.7%) and apiose (23.0 to 39.8%) were the two most
abundant components in the samples. Galactose was found in four of the seven samples:
B-lSm-IIa, B-lSm-Ilb, B-4.75h-IIa and B-4.75h-IIb. Results of GC-MS, size exclusion
chromatography and ion exchange chromatography suggested that three types of pectic
polysaccharides were isolated. They differ in sugar composition and molecular size. The
pectic polysaccharides extracted in 15 min are of two types. Samples of the first type,
B-lSm-IIa and B-lSm-IIb, were eluted with 0.2 M NaCl and samples of the second type,
A—lSm-IIIa, A-lSm-IIIb and B-lSm-III, with 0.25 M NaCl. The former have a relatively
small molecular size (the molecular weights of standard dextrans with the same mobilities
would be 15,500 and 26,000), a high neutral sugar (79.2 and 69.8%) and low'
galacturonic acid content (20.8 and 30.2%), and a relatively high ratio of
rhamnose/galacturonic acid (1/7) while the latter have a much larger molecular size (the
molecular weight of standard dextrans with the same mobility would be 205,000, 139,000
and 159,000, respectively), are more acidic (galacturonic acid content is 44.8, 67.7 and
51.3 % , respectively) and contain no galactose, and have a lower ratio of rhamnose!
galacturonic acid (the ratio for sample B—lSm—III is ll 17 and for the weighted average
of sample A-15m-IIIa and A-lSm-Illb is 1/ 16). The third type of pectic polysaccharide,
samples B-4.75h-Ila and B-4.75h-Ilb, was extracted in the 4.75 h period. They have the

smallest molecular size (the molecular weights of standard dextrans with the same

64

65
mobilities as those of samples of B-4.75h-IIa and B-4.75h-IIb are 6,120 and 12,000,

respectively), a high apiose (39.0 and 34.9% , respectively) and intermediate galacturonic
acid content (38.8 and 34.3 % , respectively) and a high ratio of rhamnose/galacturonic
acid (1/10 and 1/5, respectively).

The above results indieate that both relatively large “smooth" (assuming less
rhamnose and other neutral sugars indieates less side chains) polysaccharides and
somewhat smaller, ”hairy“ pectic polysaccharides are extracted from the cell wall of L.
minor with ammonium oxalate in 15 min. However, since it has been suggested that
pectic polysaccharides extracted with chelating reagent (such as EDTA and CDTA) in
short times are mainly from the middle lamella (68, 69) and have lower ratio of
rhamnose/galacturonic acid than those from cell walls (68), I suggest that the pectic
polysaccharides, A-15 m-IIIa, A-lSm-IIIb and B-lSm-III, were isolated from the middle
lamella (Figure 17A), while pectic polysaccharides B-4.75h-IIa and B-4.75h-IIb are
present in cell wall proper (Figure 17B). Although there are some differences, the
molecular size and rhamnose/galacturonic acid ratio of sample B-15 m-IIa and B-15 m-IIb,
are closer to those of samples B-4.75h-IIa and B-4.75h-IIb than those of samples A-15m-
IIIa, A-lSm-IIIb and B—lSm-III (Table 5), so that the samples B-lSm-IIa and B-lSm-Ilb
may also come from cell wall. Since no large size pectic polysaccharide fraction was
detected in the samples isolated with 0.05 M ammonium oxalate in the 4.75 h period, it
is possible that the time needed to extract only the pectic polysaccharides of the middle
lamella, uncontaminated with pectic polysaccharides from the cell wall proper, could be
shorter than 15 min. A proposed model for cell wall of L. minor is shown in Figure 17.

This model shows that the pectic polysaccharides of the middle lamella are different from

66

Figure 17. Proposed Model for the Primary Cell Wall of L. minor. The proposed
model of primary cell wall of L. minor shows the general structure of pectic
polysaccharides and cellulose microfibril and the linkage between them. Hemicelluloses
and cell wall proteins are not shown in the model. The pectic polysaccharides of the
middle lamella (A) are larger in molecular size, have longer segments of undisrupted
polygalacturonosyl chains (associated with Ca2 +) and less and shorter neutral sugar side
chains than the pectic polysaccharides of the cell wall (B). The key in the Figure depicts
the following: (a) cellulose microfibril, (b) polygalacturonosyl backbone, (c) two
segments of polygalacturonosyl chains associated with Ca2+ to form an "egg box", ((1)
rhamnosyl residue in the pectic polysaccharide backbones, (e) apiose and apiobiose side
chains, (1) neutral sugar side chain and (g) ester bond between pectic polysaccharides and

cellulose microfibril.

67

 

 

 

 

 

 

 

 

 

\R’

(1

MM
0000

C

13W

2

 

 

 

 

68
those of the cell wall proper.

The results show that the molecular size of a pectic polysaccharide has a greater effect
than sugar composition (e. g. the amount of galacturonosyl residues) on its ease of elution
from a DEAE-Sephadex column. Samples with a higher galacturonic acid content but
smaller molecular size, such as B-4.75h-IIa and B-4.75h-IIb, eluted from the DEAE-
Sephadex column with a lower salt concentration than did samples with a lower
galacturonic acid content but larger size, such as B-lSm-IIa and B-lSm-Ilb. In these
eases, the salt concentration required to elute a pectic polysaccharide from the DEAE-
Sephadex column is directly related to the total galacturonic acid content of the
polysaccharide. i

Apiose is more sensitive than most other sugars to hydrolysis by 2 N TFA, since
release of apiose reached a maximum about one hour at 110°C. It also was degraded
more rapidly than most of other sugars (Figure 15). A polysaccharide containing apiosyl
residues should not be hydrolyzed at a high temperatures for an extended period. If the
sample is hydrolyzed with 2 N TFA at 110°C, a good compromise time of hydrolysis
is 1.5 h.

Separation of apiitol and xylitol acetates by GC is difficult (38). Recently, I found
that when two capillary columns, SP-2380 (30 m X 0.25 mm i.d.) and DB-225 (30 m
x 0.32 mm i.d.), were used close to their maximum operating temperature (265°C and
235 °C, respectively), the apiitol and xylitol acetates were eluted separately in 9 and 18
min, respectively. Interestingly, their order of elution was reversed compared to elution
at 175°C. The separation of alditol acetates at high column temperature also has the

advantage that the quantifieation of the sugars by the integrator/computer is more precise,

69
since peaks are sharper. However, at a high column temperature, the alditol acetates with

short retention times are more likely to appear on the tail following the solvent peak and
this could interfere with quantification. Tailing, however, ean be greatly reduced by
washing the chloroform solution of the sample 5 times with water. A second benefit of
sample washing is that contaminants that could reduce column life are removed. A

disadvantage of the high column temperature is that the GC columns deteriorate more

rapidly.

10.

11.

12.

13.

14.

15.

16.

LIST OF REFERENCES

Albersheim, P. (1959) Biochem. Biophys. Res. Commun. 1, 253-256

Albersheim, P., Neukom, H. and Deuel, H. (1960) Arch. Biochem. Biophys. 90,
46—51

Aspinall, G. O. in TheBiochemistry of Plants, Stumpf, P. K. and Conn, E. E. (Ed.)
(1980) Vol. 3, 473-500

Bacic, A., Harris, P. J. and Stone, B. A. in The Biochemistry of Plants, Preiss, 1.
(Ed.), Academic Press Inc., 1988, Vol. 14, 297-371

Bacon, J. S. D. (1963) Biochem. J. 89, 103P-104P
Bacon, J. S. D. and Cheshire, M. V. (1971) Biochem. J. 124, 555-562

Bauer, W. D., Talmadge, K., Keegstra, K. and Albersheim, P. (1973) Plant
Physiol. 51, 174-187

Beck, E. (1967) Z. Hialzzmphysiol. Bd. 57, S. 444-461
Beck, E. and Kandler, O. (1965) Z. Natulji 20 B, 62-67

Bell, D. 1., Isherwood, F. A. and Hardwick, N. E. (1954) J. Chem. Soc. Part IV,
3702-3706

BeMiller, J. N. in Chemistry and Function of Pectins, Fishman, M. L. and Jen. I.
1. (Ed.), American Chemical Society, 1986, 1-21

BeMiller, J. N. (1967) Advan. Carbohydr. Chem. Biochem. 22, 25-108

Blakeney, A. B., Harris, P. J., Henry, R. J. and Stone, B. A. (1983) Carbohyd.
Res. 113, 291-299

Blumenkrantz, N and Asboe-Hansen, G. (1973) Anal. Biochem. 54, 484-489

Burgess, I. An Introduction To Plant Cell Development, Cambridge Univ. Press,
1985, 67-

Burke, D., Kaufman, P., McNeil, M. and Albersheim, P. (1974) Plant Physiol. 54,
109-115

70

17.
18.
19.

20.

21.

22.

23.

24.

26.
27.

28.

29.
30.
31.

32.

33.

35.

36.

71
Chambat, G., and Joseleau, J. -P. (1980) Carbohyd. Res. 85, C10—C12

Dabrowski, 1., Hanfland, P. and Egge, H. (1982) Methods Enzymol. 83, 69-83
Darvill, A. G., McNeil, M. and Albersheim, P. (1978) Plant Physiol. 62, 418-422

Darvill, A., McNeil, M., Albersheim, P. and Delmer, D. P. in The Biochemistry
of Plants, Tolbert, N. E. (Ed.), Aeademic Press, 1980, Vol. 1, 91-162

Dey, P. M. and Brinson, K. (1984) Advan. Carbohydr. 01cm. Biochem. 42,
265-382

Dubois, M., Gilles, K. A., Hamilton, 1. K., Rebers, P. A. and Smith, F. (1956)
Anal. Chem. 28, 350-356

Duff, R. B. (1965) Biochem. J. 94, 768-772

Eda, S. and Kato, K. (1980) Agric. Biol. Chem. 44, 2793-2801

. Epstein, L. and Derek Lamport, D. r. A. (1984) Phytochemistry 23, 1241-1246

Fry, S. C. (1986) Ann. Rev. Plant Physiol. 37, 165-186
Fry, S. C. (1982) Biochem. J. 204, 449-455

Gidley, M. J., Morris, E. R., Murray, E. 1., Powell, D. A. and Rees, D. A. (1979)
J. Chem. Soc., Chem. Commun. pp.990-992

Harris, P. J. and Hartley, R. D. (1976) Nature 269, 508-510

Hart, D. A. and Kindel, P. K. (1970) Biochem. J. 166, 569-579

Hart, D. A. and Kindel, P. K. (1970) Biochemistry 9, 2190-2196

Hang, A. and Smidsrod, O. in Solution Properties of Natural Polymers, Special
Publieation No.23, by The Chemical Society, Alden & Mowbray Ltd 1n London,
Great Britain, 1968, 273-282

Ishii, s. (1981) Phytochemistry 20, 2329-2333

. Ishii, S. (1982) Phytocherru'try 21, 778-780

Jarvis, M. C. (1984) Plant, Cell and Environment 7, 153-164

Jarvis, M. c., Forsyth, w. and Dunean, H. I. (1988) Plant Physiol. 88, 309-314

37.

38.

39.

41.

42.

43.

45.

46.

47.

48.

49.

50.

51.

52.

53.

72

Keegstra, K., Talmadge, K. W., Bauer, W. D. and Albersheim, P. (1973) Plant
Physiol. 51, 188-196

Kindel, P. K. and Cheng, L. (1990) Carbohyd. Res. 199, 55-65

Kindel, P. K. and Watson, R. R. (1973) Biochem. J. 133, 227-241

. Kiss, J. (1974) Advan. Carbohydr. Chem. Biochem. 29, 229-303

Kohn, R. and Furda, I. in Solution Properties of Natural Polymers, Special
Publieation No. 23 by The Chemicl Society, Alden & Mowbray Ltd in London,
Great Britain, 1968, 283-291

Lamport, D. r. A. (1965) Adv. Bot. Res. 2, 151-218

Lamport, D. T. A. in Frontiers of Plant Tissue Culture, Thorpe, T. A. (Ed.),
Published by The International Association for Plant Tissue Culture (1978), 235-244

. Lamport, D. T. A. in Cellulose: Structure, Modification, and Hydrolysis, Young,

R. A. and Rowell, R. M. (Ed.), John Wiley & Sons, Inc., 1986, 77-90
Lamport, D. T. A. (1970) Annu. Rev. Plant Physiol. 21, 235-270

Lamport, D. T. A. and Epstein, L. (1983) Curr. Top. Plant Biochem. Physiol. 2,
73-83

Lau, J. M., McNeil, M., Darvill, A. G. and Albersheim, P. (1985) Carbohyd. Res.
137, 111-125

Ian, J. M., McNeil, M., Darvill, A. G. and Albersheim, P. (1987) Carbohyd. Res.
168, 245-274

McNeil, M., Albersheim, P., Taiz, L. and Jones, R. L. (1975) Plant Physiol. 55,
64-68

McNeil, M., Darvill, A. G. and Albersheim, P. (1980) Plant Physiol. 66,
1128-1134

McNeil, M., Darvill, A. G. and Albersheim, P. (1982) Plant Physiol. 70,
1586-1591

McNeil, M., Darvill, A. G., Fry, 8. C. and Albersheim, P. (1984) Ann. Rev.
Biochem. 53, 625-663

Morris, E. R., Powell, D. A., Gidley, M. J. and Rees, D. A. (1982) J. Mol. Biol.
155, 507-516

54.
55.
56.
57.

58.

59.

61.

62.

63.

65.

67.

68.
69.
70.

71.

73.

73
Mort, A. J. and Bauer, w. D. (1982) J. Biol. Chem 257, 1870-1875

Mort, A. J. and Lamport, D. T. A. (1980) Analyt. Biochem. 82, 289-309
Neal, D. L. and Kindel, P. K. (1970) J. Bacteriol. 101, 910-915
Northcote, D. H. (1972) Ann. Rev. Plant Physiol. 23, 113-132

Ovodova, R. G., Vaskovsky, V. E. and Ovodov, Y. S. (1968) Carbhyd. Res. 6,
328-332

Plaschina, I. G., Braudo, E. E. and Tolstoguzov, V. B. (1978) Carbohyd. Res. 60,
1-8

. Powell, D. A., Morris, E. R., Gidley, M. J. and Rees, D. A. (1982) J. Mol. Biol.

155, 517-531

Preston, R. D. in The Physical Biology of Plant Cell Walls, Chapman & Hall,
London, 1974, 32-67

Rees, D. A. (1969) Adv. Carbohydr. Chem. 24, 267-332

Rees, D. A. (1972) Biochem. J. 126, 257-273

. Rees, D. A. (1975) MTP Int. Rev. Sci. Biochem. Ser. One, Vol. 5, 1-42

Rees, D. A. Polysaccharide Shapes, Chapman and Hall Ltd, London, 1977, 41-61

. Rees, D. A. and Welsh, E. J. (1977) Angew. Chem. Int. Ed. Engl. 16, 214-224

Rees, D. A., Morris, E. R., Thom, D. and Madden, J. K. in The Polysaccharides,
Aspinall, G. 0. (Ed.), Academic Press, 1982, Vol. 1, 195-290

Selvendran, R. R. (1985) J. Cell Sci. Suppl. 2, 51-88

Selvendran, R. R. and O’Neill, M. A. (1987) Methods Biochem. Anal. 32, 25-153
Shibuya, N. and Iwasaki, T. (1978) Agric. Biol. 01cm. 42, 2259-2266

Smith, M. M. and Hartley, R. D. (1983) Carbohyd. Res. 118, 65-80

Talmadge, K. W., Keegstra, K., Bauer, W. D., and Albersheim, P. (1973) Plant
Physiol. 51, 158-173

Tanner, G. R. and Morrison, 1. M. (1983) Phytochemistry 22, 1433-1439

74.

75.

76.

78.

81.

82.

74
Taylor, R. L. and Conrad, H. E. (1972) Biochemistry 11, 1383-1388

Thomas, J. R., Darvill, A. G. and Albersheim, P. (1989) Carbohyd. Res. 185,
279-305

Valent, B. S. and Albersheim, P. (1974) Plant Physiol. 54, 105-108

. Walkinshaw, M. D. and Amott, S. (1981) J. Mol. Biol. 153, 1055-1074

Whistler, R. L. and BeMiller, J. N. (1958) Advan. Carbohydr. Chem. 13, 289-329

. Whitrnore, F. W. (1974) Plant Physiol. 53, 728-731

. Wilder, B. M. and Albersheim, P. (1973) Plant Physiol. 51, 889-893

Williams, D. T. and Jones, J. K. N. (1964) Can. J. Chem. 42, 69-72

Yamaoka, T. and Sato, S. (1977) Bot. Mag. 90, 153-163

APPENDIX A

DERIVATION OF EQUATIONS FOR CALCULATION OF THE
AMOUNT OF NEUTRAL SUGARS AND URONIC ACID IN SAMPLES

1. BASIC EQUATION:

If the concentration of a substance in solution has a linear relationship with absorption,
the linear equation for the standard curve is:

A-b+m~c or c-— (1)

where c = the concentration of the substance
A = absorbance
b = intercept of the standard curve
m = slope of the standard curve

II. CONCENTRATION OF NEUTRAL SUGAR IN A SAMPLE MIXTURE WITH
BOTH NEUTRAL SUGAR AND URONIC ACID (8. g. pectic polysaccharide sample)

For a sample solution containing only neutral sugar, the concentration of the sugar can
be calculated by the equation:

 

a 490: .01 (2)

where c, = concentration of the neutral sugar in the undiluted sample.
A.” . = absorbance of the neutral sugar in the total sugar (phenol-sulfuric acid) test

at 490 nm.
b, and ln1 = intercept and slope of the standard curve of neutral sugar concentration vs.
absorbance for total sugar test“
D, =—fold dilution of the sample for the total sugar test

However, when both neutral sugar and uronic acid are present in the same solution,
equation (2) must be modified as follows:

-A -b
of (Am' ”0") ! D, (3)
"‘1

 

where A,” , = absorbance of total sugar in sample in total sugar test
A.,, , = absorbance of uronic acid in total sugar test
A.,, , cannot be measured directly in the sample. But it can be calculated from:

75

76

e
A490,'b2+’"2'3u (4)
1

where b2 and 1112 = intercept and slope of the standard curve of uronic acid concentration
vs. absorbance in the total sugar test.
c,I = concentration of uronic acid in the sample.

The concentration of uronic acid in the sample is:

- )4,
c.-“”“' A... .02 (5)
"‘3
where A,,,, and A,,,, = absorbance of total and neutral sugar, respectively, in the sample in
the uronic acid test
b, and m, = intercept and slope of the standard curve of uronic acid concentration

vs. absorbance in the uronic acid test
D2 = —fold dilution of the sample for the uronic acid test

A5,o , cannot be determined directly from the sample. But it can be calculated from:

c
A,,(,u-b,,+m4 31 (5)
2

where b, and m, = intercept and slope of the standard curve of neutral sugar vs.
absorbance in the uronic acid test

The expression for A5,, ,, in (6) is substituted into (5) to give:

e
Am-b-rn-J-b
, 4 4D: 3

c - -D
It m3 2

 

The expression for c, in (4) is substituted into (7) to give:

C
A -b —m -—"-b

"'3 Dr

 

A490,'b2*’"2‘
The expression for A,” , in (3) is substituted into (8) to give:

C
3 1
c; D,
"'1

 

 

After being simplified, the equation (9) becomes:

77

A __ "'2 .D2
60' "'3'Dl

 

(A520 ‘bs'bJ "br‘bz
r . D1

1_ rnz'm4 "'1
"'1 an,

 

C-

(10)

*If the neutral sugar component in a pectic polysaccharide sample is mostly galactose,
then b, and m, can be consider to be the intercept and slope of the standard curve of galactose
concentration vs. absorbance in the total sugar test, and the concentration of the galactose can
be obtained by equation (10). Since the apiosyl residue in a reduced pectic polysaccharides
of L. minor is also a significant neutral sugar component (together with the galactosyl residue)
and it has an absorbance response in the total sugar test quite different from that of galactose,
b1 and 111, must be modified as follows:

If x is the portion of the neutral sugar component of a sample that is apiose, then

A -b
get-flip, (11)
"'5
where c, and D1 are the same as in equation (2) above
A.,,0 ,, = absorbance of apiose of the polysaccharide sample in the total sugar test
b, and m, = intercept and slope of the standard curve of apiose concentration vs. absorbance
in the total sugar test

Similarly, the concentration of galactosyl residue in total neutral sugar components can be
calculated by equation:

A -b
cu(1.x)-_”_°mL‘.Dl (12)
6

where A.,, u = absorbance of galactose of the polysaccharide sample by the total sugar test
b6 and m, = intercept and slope of the standard curve of galactose concentration vs. its
absorbance in the total sugar test

Equation (11) and (12) can be re-written as:

c c '
.1 Dr a! D1
They are combined to give:

cum, x+c'1n‘fl-x)
AmflrrAm“ Dr +b, +b,s

 

Rearranging gives:

78

.. (AderAm') -(b5+b‘) ~D
rn,-x+rn‘(l —x)

Cu

 

1

(13)
Compare equation (13) with equation (2). Since A.,,0 ,, = A.,“, + A.,, a, then:
m=m+m 00
mi = ms?! + mail-X) (15)
Similarly, in equation (6) it follows that:
b. = bi + be (16)
m4 = myx + m,-(1-x) (17)

where b, and m, are the same as in equation (6),

b-, and m, = intercept and slope of the standard curve of apiosyl residue concentration
vs. absorbance in the uronic acid test

b, and m, = intercept and slope of the standard curve of galactosyl residue concentration
vs. absorbance in the uronic acid test

By using the equations (14), (15), (16) and (17) to ealculate b,, m,, b, and m, and
substituting these in the equation (10), the concentration of neutral sugar in a sample
(including apiosyl and galactosyl residues) ean be obtained.

III. CALCULATION OF THE CONCENTRATION OF URONIC ACID IN A
SAMPLE MIXTURE WITH BOTH NEUTRAL SUGAR AND URONIC ACID

The equation (7) above is:

c
Am,'b4'm43: “b3 (7)
c; D,
"‘3
where A,,, , is the experimentally measured absorbance (used in equation(5)) and 8,, b,,
a,, b, and D, are as described above. When the concentration of neutral sugar, c,” is
obtained from equation (10), it can be used in equation (7) to calculate the concentration

of uronic acid, cm in the sample.

 

APPENDIXB

DERIVATION OF THE EQUATION USED TO CALCULATE THE

GALACTITOL ACETATE-D, IN GALACTITOL ACETATE

For a nominal ion x in the mass spectrum of galactitol acetate-D,, ions with five

different masses should be found in the typical ion cluster of related ions. These are
shown in the diagram:

 

 

 

it I.
1:42 " Inn.) + I1142(0)
T
r i
0 :
N :
I E Lump)
N :
T :
E :
1g :
I 1"” . 1x43 " 11.1-ammo.)
Y i v
.. '7 f i 111:4 " Lam-2(a)
[14mg .
..... l -

 

 

 

I .
x x+1 x+2 x+3 x+4

where x and Ix = the nominal ion x and its intensity, respectively.
x+1 and I”, = the ion with one mass unit larger than ion x and the intensity ofthis
ion, respectively. The additional one mass unit is mainly due to the
presence of natural 2H, 13C or ‘70 in x.
x+2 and I“, = the ion with two mass units larger than ion x and the intensity of this
ion, respectively. The intensity of the ion is due in part to the

presence of two 13C atoms or an “O atom in ion x (labeled 1”,“,

79

80
where 2(n) means the two mass units contributed by natural

isotopes), and in part to two introduced deuterium atoms in ion x
(Ix-mm)- '

x+3 and I“, = the ion with three mass units larger than ion x and the intensity of
this ion, respectively. The three mass units are mainly contributed
by two introduced deuterium atoms and one 2H, 13C or 1"’O atom (1b.,
= I:+2(D)+1(n))-

x+4 and I”, = the ion with four mass units larger than ion x and the intensity of this
ion, respectively. The additional four mass units are mainly
contributed by two introduced deuterium atoms and two 13C atoms

or an 18O atom (Ix+4 = Lama“).

The sum of the intensities from all ions with two introduced deuterium atoms is:
(Ln-2n» + Ix+2(n)+l(n) + Ix+2(D)+2(n)) = (Inna) + Ix+3 + I...)

Since galactitol acetate-D, has a symmetrieal structure (not considering the deuterium
atoms), then for any type of ion with two introduced deuterium atoms there must be a
corresponding ion with the same structure except without the introduced deuterium
atoms. Therefore, in above ion cluster, the total intensity of the ioris from galactitol
acetate-D, is: 2 X (1",0, + I”, + 1",).

When the ratio (R) of galactitol acetate-Dflgalactitol acetate-D, + galactitol acetate)
is 1 then the galactose content of the sample is zero. R ean be ealculated as follows:

,_ 2x<I.....,+I...+I..>

Ix+lx+1 +Ix+2 +Ix+3+Ix+4

(1)

 

where, 1“,” = ,,, - 1”,” 1",“, can not be obtained directly but can be ealculated

81
from: I,,,,,, = I, x F,,, (2)

where F,,, is the expected abundance ratio of ion x+2 ion due to the occurrence of two
natural 13C atoms or an 18O atom in ion x. Therefore, 1“,“), can be calculated as follows:

Ix+2(D) = 1m - (I. X Fm) (3)
Equation (1) ean now be rewritten as:

R- 2x[(Ix+2-IxXFx,2) +1164.li X H” (4)
If" +1 +1 +1 98

n+1 x+2 x+3 x+4

 

where 100/98 accounts for NaBD, being 98% deuterium. The values for II and I“, were
taken from the mass spectrum while L“, IM., and L,, were calculated by using the
expected abundance ratios, F,“ and F“, rather than taking the values from the mass
spectrum:

1m = 1. >< Fm

I... = 1....» x F... = (I... - I. x F...) x F...

1144 = 114-20)) X F3+2 = (Ix-+2 ' Ix X Fir-+2) X Fx+2

Substituting, equation (4) becomes:

R- 2X[(I,,,-Ixxe+7) +(Ix+2-IxXFx+2) XFx+1+(Ix+2-IxXFx+2) XFxJ] X 100
[x+(IxXF n+1) +Ix+2+(lx+2-IxXFx+2) XF x+1+(Ix+2—IxXF :42) Xde 98

 

which ean be simplified to:
R- zxaxd-IxXFx-rz) x 100 (5)
Ixx(1 -Fm) +1”,2 98

When the ion pair 217/219 mlz was chosen to represent galactitol acetate/galactitol
acetate-D,, I, = intensity of ion 217 m/z and I“, = intensity of ion 219 mlz. Ion 217
contains 13 H, 9 C and 6 0, so abundance ratios, Fm and F,,,, can be calculated as

follows:

82

 

 

 

 

Ion x x+1 x+2

Natural D in 13 H 0.01x13 = 0.13

"C in 9 C 1.1x9 = 9.9 0.006x(9)2 = 0.486
r’O in 6 O 0.04x6 = 0.24

"0 in 6 O 0.2x6 = 1.2

A (abundance? 100 A,,,=0. 13 +9.9+0.24=10.27 A,,,=0.486+1.2=l.686
F“ 1 F,,,=0. 1027 F,,,=0.01686

 

 

 

 

 

 

* A represents the relative abundance of an ion.
** F (factor) represents the abundance ratio of an ion. F,“ = Ah/ 100 when the
abundance of ion x is set equal to 100.

Similarly, when the ion pair 259/261 mlz was chosen to represent galactitol
acetate/galactitol acetate-D,, Ix = intensity of ion 259 mlz and I“, = intensity of ion 261

mlz. Ion 259 contains 15 H, 11 C and 7 0, so abundance ratios, Fm and F,,, can be

 

 

 

 

 

 

 

 

ealculated as follows:
Ion x x+1 x+2
NaturalDinlSH 0.01><15 = 0.15
"C in 11 C 1.1x11 = 12.1 0.006><(11)2 = 0.726
"O in 7 O 0.04x7 = 0.28
"o in 7 o 0.2x7 = 1.4
A 100 A,,,=0.15+12.1+0.28=12.53 A,,,=0.486+1.2=2. 126
F 1 F,,,=0. 1253 F,.,-—-0.02126

 

 

APPENDIX C

83

Carbohydrate Research, I99 (I990) 55—65 55
Elsevier Science Publishers B.V., Amsterdam — Printed in The Netherlands

The acetylation of apiitol in the determination of apiose

Paul K. Kindel and Liang Cheng
Department of Biochemistry. Michigan State University. East Lansing, Michigan 48824 ( U.S.A.)

(Received July 10th, l989; accepted for publication October lSth, [989)

ABSTRACT

The complete acetylation of apiitol required 9 h when acetic anhydride at 120° was used and sodium
acetate was the eetalyst. Both apiitol pcntaaoctatc and apiitol tctraacctate were detected before acetylation
was complete. When the reaction was done in dimethyl sulfoxidc, with l-methylimidazolc as the catalyst, a
third compound was observed, and identified as l,2,4-tri-0-acetyl-3-C-(acetoxymethyl)-3-0-(methylthio-
methyl)-D-glycero-tctritol [3-0-(mcthylthiomcthyl)apiitol tetraacctate] by gas-liquid chromatography and
mass Spectrometry. ln N,N-dimethylformamidc, with l-methylimidazole as catalyst, the acetylation of
apiitol was essentially complete in 4 h at 85°, and the formation of methylthiomcthyl ether was avoided. A
method for preparing alditol acetates using l-mcthylimidazole as the catalyst, and suitable for samples
containing apiose as well as ordinary sugars, is described. The separation of apiitol pentaacetate from xylitol
pentaacctate by gas-liquid chromatography proved difficult. However, a virtually complete separation of
the peracctates of apiitol and xylitol as well as complete separation of those of rhamnitol, fucitol. arabinitol,
mannitol, galactitol, glucitol, and myo-inositol, plus apiitol tetraacctatc and 3-0-(methylthiomethyl)apiitol
tetraacctate, was accomplished with a 30 m x 0.53 mm (i.d.) SP-2380 column in 49 min, and on a 30 m x
0.75 mm (i.d.) SP-2330 column in 82 min. A complete separation of apiitol and xylitol pentaacetates as well
as four other alditol peracetates was obtained with a 60 m DB-l column in 15.2 min, however this column did
not resolve the acetates of fucitol and arabinitol. A variety of other columns and column conditions were
ineffective.

lNTRODUCTlON

Apiose [3-C-(hydroxymethyl)-D-glycero-aldotetrose] is widely distributed in the
plant kingdom'. It is present both in cell-wall polysaccharides and in non-cell-wall
compoundsz. It is therefore of interest when the sugar composition of plant cell walls
and cell-wall fractions is determined, typically by the preparation of alditol acetates
after acid hydrolysis of the sample. All of the known sugars present in plant cell-wall
material, except apiose and aceric acid’, have only primary and secondary hydroxyl
groups, and consequently their alditols are acetylated relatively easily. Apiose and
aceric acid, on the other hand, have tertiary hydroxyl groups, which past experience has
shown are difficult to acetylate when the more common catalysts such as pyridine are
used‘. The conditions for complete acetylation of apiitol have not been established,
although the assumption may have been made that the published conditions for the
acetylation of other alditols also gave complete acetylation of apiitol. We investigated
the acetylation of apiitol by two methods currently used to acetylate alditols”, and
found that complete acetylation in both cases required considerably more rigorous

0008-62l5/90/$ 03.50 © [990 Elsevier Science Publishers B.V.

84

56 P. K. KINDEL, L. CHENG

conditions than those needed for sugar alcohols having only primary and secondary
hydroxyl groups. We also found that when the procedure of Blakeney et al.7 was used
under conditions that gave complete conversion of apiitol, substantial amounts of
l,2,4-tri-0-acetyl-3-C-(acetoxymethyl)-3-0-(methylthiomethyl)-D-glycero-tetritol
[herein called 3-0-(methylthiomethyl)apiitol tetraacetate] were formed as a side-prod-
uct.

EXPERIMENTAL

Materials. — Apiose was isolated as described previously“. Other sugars and
myo-inositol were obtained from Pfanstiehl Laboratories, Inc. l-Methylimidazole,
4-dimethylamin0pyridine, dimethyl sulfoxide, and fully deuterated acetic anhydride
were obtained from Aldrich Chemical Co., Inc. N,N-Dimethylformamide was pur-
chased from Sigma Chemical Co. myo-Inositol hexaacetate (m.p. 2l3—214° was pre-
pared by a modification of the procedure described by Wachowiak and Conners" for
glucose pentaacetate. Columns packed with SP-2330, SP-2380, and DB-l were pur-
chased from Supelco, Inc. and J & W Scientific, Inc., respectively.

Preparation of alditol acetates. — Alditol acetates were prepared by the following
four procedures: (I) Albersheim et al.’, (2) Blakeney et al.“, (3) Steglich and H6fie9, and
(4) a modification of the prowdure of Blakeney et al."'7 The particular procedure used,
along with any changes, is stated when the results of individual experiments are
described. In the preparation of individual samples, 0.89—1.78 pmol of apiose and
0.82-l .0 pmol of rhamnose, fucose, arabinose, xylose, galactose, glucose, and mannose
were used“. Except where indicated otherwise, myo-inositol was added to the samples as
an internal standard.

The procedure of Blakeney et al.7 was modified as follows so that complete
acetylation of apiitol was achieved without formation of any side-product (acetylation
procedure IV). The sample of apiose was in 0.2 mL of M ammonium hydroxide. and
after 1 mL of 2% (w/v) NaBH,, in DMF was added it was kept for 60 min at 40°, cooled
to 22°, acidified with 0.2 mL of glacial acetic acid, and mixed. For acetylation, 0.2 mL of
l-methylimidazole and 2 mL of acetic anhydride were added and the sample was mixed
and heated for 4 h at 85°. Five mL of water was added to the solution at 22° and after
mixing and cooling to 22°, 1 mL of dichloromethane was added. The sample was mixed,
then centrifuged, and the organic phase used for gas chromatography. The modified
procedure was also tested with rhamnose, fucose, arabinose, xylose, mannose, galac-
tose, glucose, and m yo-inositol.

Gas chromatography. — Gas-liquid chromatography of alditol acetates was
performed initially with a Varian Aerograph, Series 2100, and more recently with a
Varian Model 3700 instrument, each equipped with a flame-ionization detector. A

‘Abbreviations: e.i., electron impact: c.i.. chemical ionization: M‘. molecular ion; DMSO. dimethyl
sulfoxide; DM F, N,N-dimethylformamide; all sugars have the D configuration except for L-rhamnose.
L—fucose, and L-arabinose.

85

ACETYLATION 0F APIITOL 57

capillary glass column (30 m x 0.75 mm, i.d.; film thickness 0.2 pm), coated with
SP-2330 and a fused-silica column (30 m x 0.53 mm, i.d.; film thickness 0.2 pm), coated
with SP-2380, were used. Samples were injected directly into the SP—2330 column. When
the SP-2380 column was used the Model 3700 chromatograph was equipped with a
splitter, with the split ratio ranging from 1:7 to 1:15. With the SP-2330 column, nitrogen
was used as both the carrier and makeup gas, usually at flow rates of 5 and 40 mL. min ‘ ',
respectively, while with the SP-2380 column helium was the carrier gas and the flow rate
normally was 5 mL.min ‘ '. Data from these two columns were collected with a Hewlett—
Packard Reporting Integrator, Model 3390A. Alditol acetates were also separated on a
DB-l fused-silica capillary column (60 m x 0.25 mm, i.d.; film thickness 0.1 pm)
attached to a Hewlett-Packard gas chromatograph, Model 5840A, equipped with a
splitter, a flame—ionization detector, and a 5840A data terminal. Helium was the carrier
gas and the flow rate was 0.8 mL.min". The split ratio usually was 1:14.

Samples injected into gas chromatographs ranged in size from 0.1 to 1.0 [1L when
no splitter was used and from 1 to 2.5 pL with a Splitter.

Mass spectrometry. — EL and c.i. mass spectrometry were performed with a
J EOL HXl 10-1-1 F, double-focusing mass spectrometer interfaced to a Hewlett—Pack-
ard gas chromatograph, Model 5890A, which was equipped with a splitless injector and
a 08-1 fused silica column (15 m x 0.53 mm, i.d.; film thickness 1.5 pm). Helium was
the carrier and makeup gas; the flow rates were 10 and 20 mL.min“ ', respectively. The
ion source temperature was 200° and the ionizing voltage was 70 eV. For c.i. mass

spectrometry, ammonia was the reagent gas.
RESULTS

Acetylation with sodium acetate as catalyst. —— Complete acetylation of apiitol
required 9 h when acetic anhydride at 120° was used and sodium acetate was the
catalysts. Samples prepared from apiose by the procedure of Albersheim et al.’ were
analyzed with the 30 m SP-2330 column at 200° (see Experimental) at acetylation times
of 20 min, 40 min, 1, 1.5, 2, 3, 4, 5, 7, and 9 h. The observed values for apiitol
pentaacetate and apiitol tetraacetate were 1074] .4, 59.5, 69.3, 72.3, 85.7, 98.0. 97.0,
99.2, and 100% and 89.3, 58.6, 40.5, 30.7, 27.7, 14.3, 2.0, 2.9, 0.8, and 0%, respectively,
calculated from the area under each chromatographic peak, with the sum of each pair of
areas set equal to 100%. The acetates were identified by gas—liquid chromatography and
mass spectrometry (Figs. 1a and 1b). Based on the fragmentations proposed, ions at m/z
289, 303, 319, 347, and 362 would be unique to the pentaacetate. Of these only m/z 289
was present in substantial amounts; it had a relative intensity of 16.8 with respect to the
base peak“ at m/z 187. The other ions were not present except for m/z 303, having
intensity 0.82. In the spectrum of the tetraacetate derivative all these ions were un-

‘In this paper: (1) the base ion (peak) was assigned a value of 100. (2) relative intensities are with respect to
the intensity of the base ion. and (3) ions with m/: of 50 or less were not considered.

86

 

 

 

 

 

 

 

 

 

 

 

58 P. K. KINDEL, L. CHENG
3' "7 941362) 31:98:13?”
us ‘182
'° ’ a
g 2174:4347) '-0
'g ‘° UO-lflcs—SOB
.D
a
,. ,0, “OJ m “‘13:"*2°9Hc0~: "5:...23
“'1 1th
a: 127 Wad 289-024H187-Ol45-9103
n :05»
2° "5 s 209 } Fifi, “'
.8 '5. 2” 2‘7 ?
3 L L 2:0 (“MW-.115
o- rfi-:fi,e-!+,k- .,A-e (:113
so 100 150 200 300
at:
in
b. 187 . :200-0140-9 98
‘128
”4 ’ OH
21740-1305) 3.00
8 w 0
g l ate-must! 2‘!
4' 854- 31874924? "m“ '45_.‘m
i 404 1‘5 175 ' Fifi“ ”w 247:l:—OH5—O|03
r: outs (2611-le
20‘ 217 , 4’1
0
11113360 (3051-0217-0115
JJ 4.119.247 r—f—'
o ......... liege}: -ea--~i.---v- 0';
80 200 280 300

Fig. l. E.i. mass spectra of (a) apiitol pentaacetate, and (b) apiitol tetraacetate. See Experimental for
procedure. Parentheses mean the ion was not detected. Double-headed arrows represent two or more steps.
Sequences to minor ions are not shown.

detectable or scarcely apparent. Ion 289 therefore can be used to distinguish between
apiitol pentaacetate and apiitol tetraacetate.

According to the fragmentations proposed, the ions at m/z 133, 159, 218, 261, 277,
305, and 320 are unique to the tetraacetate. Of these only 133 and 218 were present in the
spectrum of the tetraacetate at intensities greater than 0.85; their peak heights were 5.3
and 2.3, respectively. In the spectrum of the pentaacetate, the ions at m/z 133 and 218
had relative intensities of 2.8 and 3.8; the remaining ions were of intensity 0.38 or less
except for m/z 159 at 2.2. The relative intensity of ion 133 was always 2—3 times greater in
the spectra of the tetraacetate than in the spectra of the pentaacetate. The reactions
proposed to account for the spectra of the two derivatives accounted for 92 and 91%,
respectively, of the total ions detected (background ions included, 3 0.05% of the total
ion current).

Acetylation with I-methylimidazole as catalyst — When apiitol and acetic an-
hydride were heated at 35° in the presence of l-methylimidazole and DMSO for various

87

 

 

 

 

 

 

 

 

 

 

 

ACETYLATION or APIITOL 59

(I)

g 70’ 0‘ b s

.5

S 60

L

m as

U 50 a in 2L

3; 3

‘540 :3

° 4

E 30 - ___._ r8, 1

9 2

B 20 r- 8

*c‘

a) 10 b

E ..__—i

a) o a l A 1 l l l l i J a I

n. o I 2 3 4 6 8 10 12 14 16 18 20 O 5 10 15
Time of acetylation (h) Time (min)

Fig. 2. Acetylation of apiose by the procedure cf Blakeney et all Samples were prepared from apiose as
described’, except that acetylation was performed at 35° for the times indicated. A single extraction with
dichloromethane was performed, resulting in underestimation of the proportions of apiitol tetraacetate (see
Results). Samples were analyzed with the 60 m DB-l column as described in the Experimental part. Column
operation was isothermal at 200°. (3) Time course of formation of products, showing apiitol tetraacetate (o),
apiitol pentaacetate (o). and 3-0-(methy1thiomethy1)apiitol tetraacetate (A) as percentages of the total
dichloromethane-extracted material at each time point. (b) Gas-liquid chromatogram of the products.
Acetylation time for this sample was 5 h, and xylitol pentaacetate in dichloromethane was added to the
mixture before g.1.c. Compounds 1, 2, 3, and 4 are apiitol tetraacetate, xylitol pentaacetate, apiitol
pentaacetate, and 3-0—(methylthiomethyl)apiitol tetraacetate, respectively.

time periods the results shown in Fig. 2a were obtained. A typical separation of the three
compounds detected is shown in Fig. 2b. Two of the compounds were identified as
apiitol pentaacetate and apiitol tetraacetate by their c.i. mass spectra. These were as
shown in Figs. 1a and lb, with the ions at m/z 289 and 133 having the same relative
intensities in the respective spectra. The c.i. mass spectrum of the compound considered
to be apiitol pentaacetate showed ions at m/z 380 (M + NH,*), 363 (M + H‘), and 303;
their relative intensities were 100, 0.31, and 55, respectively. When fully deuterated
acetic anhydride was used, ions at m/z 395 (M + NH,*), 378 (M + H‘), and 315 had
relative intensities 28.5, 0.2, and 100, respectively. The c.i. mass spectrum of the
compound considered to be apiitol tetraacetate showed ions at m/z 338 (M + N113),
321 (M + 11*), and 303 at relative intensities 62, 2.7, and 100, respectively. Fully
deuterated acetic anhydride gave ions at m/z 350 (M + NH,+), 333 (M + H“), and 315
with intensities 7.5, 0.84, and 100. These data confirm the identity of two of the products
(Fig. 2a) as apiitol pentaacetate and apiitol tetraacetate, as well as provide confirmation
of the identity of the two compounds formed from apiose when the procedure of
Albersheim et al.’ was used.

The third compound in Fig. 2a was identified as 3-0-(methylthiomethyl)apiitol
tetraacetate by mass spectrometry. The base ion in the c.i. spectrum (Fig. 3) had m/z of
61 and probably was mainly CH,SCH2*. The fragmentations proposed accounted for
87.5% of the total ions in the spectrum (background included). The c.i. mass spectrum
of the compound showed ions at m/z 398 (M + NHf), 381 (M + H+), and 303 at
relative intensities of 100, 0.38, and 55, respectively. Fully deuterated acetic anhydride

88

 

 

 

 

 

 

 

 

 

60 P. K. KINDEL, L. CHENG
100
5‘ 1413801
no
.31
0 I365) $00
g “4 3°3 . 0 too-ow:
§ 81 m 81.-MGM" c"!
“1 14500307 3230'3-0
.E .0 23! I—1—' 31:3?
5.: mice-e 04:3,}:
2m 2“ (:21 201-010-099
,0, 9 m _ an”: ea,
'03 ‘5 8“ 1 l 9'0 (3851
., A .. - 1 A
to m .5. 200 250 .3. 3;. °'=

Fig. 3. E.i. mass spectrum of 3-0-(methylthiomethyl)apiitol tetraacetate. See legend to Fig. 1.

gave ions at m/z 410 (M + NH, *), 393 (M + H *), and 315, having intensities 24.8, 0.47,
and 76.2, respectively. Ion 61 was now the base ion. In addition, the following data
further support the identification of the third compound: (1) the compound was not
formed when apiitol was acetylated by procedure 1V (information presented later), (2)
the data in Fig. 2a show that formation of the compound ceased when apiitol tetraace-
tate was no longer present. (3) the compound was not formed when the other tested
acetylation procedures were used with apiitol, and (4) it was not formed when alditol
peracetates of rhamnose, fucose, arabinose, xylose, mannose, galactose, and glucose
were prepared by the procedure of Blakeney et al.7, either as described in the reference or
by acetylation for 4 h at 85°.

Acetylation with 4-dimethylaminopyridine as catalyst. — When apiitol was acety-
lated by acetic anhydride with 4-dimethylaminopyridine as the catalyst”, only apiitol
pentaacetate and apiitol tetraacetate were observed (data not presented). Under our
best conditions (9.5 h, 22°. no pyridine added), 96% of the material detected by the gas
chromatograph was apiitol pentaacetate, and 4% was apiitol tetraacetate. The yield of
apiitol pentaacetate was not increased by acetylating at 40°, increasing the acetylation
time to 24 h, adding pyridine together with 4-dimethylaminopyridine, or various
combinations of these.

Acetylation procedure for apiose and other sugars — The modification of the
procedure of Blakeney et al.7 permitting its use with apiose consisted of replacing
DMSO with DMF and increasing the time and temperature of acetylation to 4 h and
85°, respectively. With the new procedure (acetylation procedure IV), greater than 99%
of the apiitol recovered was fully acetylated and no side-products were observed. The
time course is given in Fig. 4. A small amount of apiitol tetraacetate was still present
after 4 h of acetylation at 85° (0.56%) and some was present even after 18 h at 85°
(0.17%). However, since these amounts are less than 1%, acetylation for 4 h at 85°
should be acceptable for virtually all work.

The degree of reduction of sugars treated with NaBH, in DMF was examined

89

ACETYLATION OF APIITOL 61

 

58888833888

 

 

Percent of total apiitol derivatives

1 l

65 I 2 3
Time of acetylation (h)

 

O

0
ho-

Fig. 4. Formation of apiitol pentaacetate from apiitol and acetic anhydride with acetylation procedure IV.
Samples were prepared as described in the Experimental part, except that acetylation times were varied. A
single extraction with dichloromethane was performed, resulting in underestimation of the apiitol tetraace-
tate. Samples were analyzed with the SP-2380 column operating isothermally at 200°. The curves show
apiitol tetraacetate (o) and apiitol pentaacetate (c).

after 15, 30, 60, and 90 min at 40°. A sample was prepared that contained the seven
sugars listed in the Experimental but no apiose; a second contained only apiose. Sugars
were reduced and subjected to acetylation procedure 1V, except that the reduction time
was varied as indicated above and the sample containing the seven sugars was acetylated
for 10 min at 22°. myo-Inositol hexaacetate (0.15 mg) was added with the dichloro-
methane. The relative peak areas of the peracetates of apiitol, xylitol, mannitol,
galactitol, and glucitol with respect to myo-inositol hexaacetate were maximal at 15 min
of reduction and did not change more than 14% between 15 and 90 min. The relative
peak areas of the peracetates of rhamnitol, fucitol, and arabinitol at 15 min were 93, 94,
and 93%, respectively, of those obtained at 90 min. The relative peak areas of the
peracetates of rhamnitol, fucitol, arabinitol, and the other alditols were not different by
more than i4% at 60 and 90 min. These results showed reduction in DMF proceeded
more rapidly than in DMSO’. '

When the acetylation mixture of Blakeney et al.7 was held at 22° for extended
periods or heated at 85°, the solution darkened'o owing to a reaction between 1-
methylimidazole and acetic anhydride. When 0.2 mL of l-methylimidazole, 2 mL of
acetic anhydride, and 1 mL of DM F were heated for 4 h at 85° with 0.1, 0.15, or 0.2 mL
of M NH,OH neutralized with glacial acetic acid, the absorbance of the solutions at 420
nm was 1 .85, 0.47, and 0.29, respectively. Color formation also decreased when decreas-
ing amounts of l-methylimidazole were used. Addition of mannitol (a representative
alditol) and m yo-inositol together to the acetylating solution or substituting water for
the neutralized ammonium hydroxide solution had no significant effect on color
formation.

Completeness of extraction by dichloromethane was examined. Apiose was
treated by acetylation procedure IV. except that the sample was heated only 15 min at

62 P. K. KINDEL, L. CHENG

85°. A mixture of rhamnose, fucose, arabinose, xylose, mannose, galactose, and glucose
was subjected to acetylation procedure IV as described. Samples were prepared in
duplicate. myo-Inositol hexaacetate (0.15 mg) was added with the first 1.0 mL of
dichloromethane. Each sample was extracted successively four times, each time with 1.0
mL of dichloromethane, and an aliquot of each organic phase was analyzed by gas
chromatography. The recoveries, based on total amount extracted, of compounds
present in organic phases 1 through 4, respectively, were: (1) apiitol tetraacetate, 38, 40,
16, and 5.8%, (2) apiitol pentaacetate, 81, 19, 0, and 0%, and (3) myo-inositol hexaace-
tate, 86, 13, 1.0, and 0.2%. The recoveries of the peracetates of rhamnitol, fucitol,
arabinitol, xylitol, mannitol, galactitol, and glucitol in organic phases 1 through 3
ranged from 78 to 87, 13 to 21, and 0 to 3%, respectively, and were zero in organic phase
4. Clearly, if apiose is incompletely acetylated and a single extraction with dichloro-
methane is performed, the amount of apiose originally present in the sample will be
seriously underestimated.

When acetylation procedure IV was used with rhamnose, fucose, arabinose,
xylose, mannose, galactose, and glucose, the yields of alditol acetates ranged from 104
to 1 19% of those obtained with the original procedure’. When acetylation procedure IV
was used, but with a reaction time of 10 min at 22°, the yields of the individual alditol
acetates were basically the same as from the original procedure’.

Separation of alditol acetates. — Complete separation of apiitol and xylitol
pentaacetates was achieved on a 60 m DB-l capillary column (Fig. 2b). When the DB-l
column was used with a temperature program designed for rapid elution, apiitol
pentaacetate, apiitol tetraacetate, myo-inositol hexaacetate, and the alditol peracetates
of rhamnose, xylose, mannose, glucose, and galactose were separated in 15.15 min. The
acetates of fucitol and arabinitol were separated from the others but not from each
other, and mannitol hexaacetate was not separated from 3-0-(methylthiomethyl)apiitol

 

68011

 

 

 

 

Detector response

 

 

 

 

 

 

Jg
(J 1 l 1 J l A l l 1 1 4L
0 510152025303540455055

Time (min)

Fig. 5. Gas-liquid chromatogram showing the separation of alditol peracetates and related compounds.
The sample was prepared by the procedure of Blakeney et al.’. except that acetylation was for 15 min at 85°.
The sample was analyzed with the SP—2380 column programmed to hold at 170° for 32 min, then increase 4°.
min " to 210°, and hold at 210°. Peaks represent the peracetates of the listed alditols. and other compounds
as noted: 1, rhamnitol; 2, fucitol; 3, arabinitol: 4. apiitol; 5. xylitol; 6, mannitol; 7, apiitol tetraacetate; 8,
galactitol; 9. 3-0-(methylthiomethyl)apiitol tetraacetate; 10. glucitol; 11. myo-inositol.

 

91

ACETYLATION OF APIITOL 63

tetraacetate. Virtually complete separation ofapiitol and xylitol pentaacetates as well as
separation of the alditol peracetates of the other six sugars listed above, apiitol tetraace-
tate, 3-0-(methylthiomethyl)apiitol tetraacetate, and myooinositol hexaacetate was
achieved in 49 min with the SP-2380 column (Fig. 5). A similar separation of these
compounds was achieved in 80 min with the SP-2330 column. The order of elution of the
compounds from the latter column was the same as that shown in Fig. 5, except that
apiitol tetraacetate eluted immediately before mannitol hexaacetate rather than after.
The use of a variety of other carrier-gas flow rates and temperatures with these two
columns did not improve the separation of apiitol and xylitol pentaacetates. A number
of other columns, both packed and capillary, were tested but gave either no or only
partial separation of apiitol and xylitol pentaacetates.

DISCUSSION

Complete acetylation of apiitol is considerably more difficult to achieve than
complete acetylation of alditols having only primary and secondary hydroxyl groups.
This is shown by the requirement for a 9-hour reaction period at 120° when the
procedure of Albersheim et al.’ is applied to apiitol.

The procedure described by Blakeney et al.7 is not suitable for the acetylation of
either apiitol or presumably any other alditol possessing a tertiary hydroxyl group
because of formation of a methylthiomethyl-ether derivative as a side-product. This
might be expected, because the acetylation mixture is analogous to reaction mixtures
containing DSMO and acetic anhydride that are used in organic synthesis to oxidize
primary and secondary alcohols to the corresponding carbonyl compounds"‘”. A
common side-product in these oxidations is the methylthiomethyl ether of the alco-
hol' "'2. In the usual applications of the Blakeney et al. procedure acetylation evidently
proceeds much more rapidly than oxidation, consequently oxidized sugars have not
been observed. When tertiary alcohols are subjected to these conditions oxidation
cannot occur, allowing methylthiomethyl ether formation to proceed, often in high
yield""‘. Moreover, the formation of the methylthiomethyl ether is facilitated by acetic
acid”, which is present in substantial amounts in the Blakeney et al.’ acetylation
mixture. We found that heating the acetylation mixture for 4 h at 85° resulted in an
organic phase 3-fold larger in volume than those obtained from samples kept at 22° for
10 min or 4 h, perhaps because of the formation of acetoxymethyl methyl sulfide by
Pummerer rearrangement" of acetoxydimethylsulfonium acetate derived from the
DMSO and acetic anhydride present.

Color formation in the acetylation solution can be diminished by increasing the
volume of the aqueous sample solution or by decreasing the amount of l-methylimida-
zole used, or both. However, both changes have potential for decreasing the yield of
alditol acetates. The quantities specified in acetylation procedure IV provide for a full
yield of alditol acetates while minimizing color formation. Moreover, this procedure
should be suitable for alditols possessing tertiary hydroxyl groups.

The mass spectra presented in Figs. la, 1b, and 3 can be explained in large part by

92

64 P. K. KINDEL, L. CHENG

reactions reported in the literature. Particular relevant are processes resulting in the loss
of acetic acid, ketene, formaldehyde, acetic anhydride, and water'°"7, but other reac-
tions are also involved. In this connection it is of interest to consider the ion of m/z 244,
found in the spectrum of 3-0-(methylthiomethyl)apiitol tetraacetate, which we suggest
must be formed from m/z 303. The change involves conversion of a cation having an
even number of electrons (m/z 303) into an odd-electron cation-radical (m/z 244) and a
radical (m 59), a process that is usually considered energetically unfavorable. However it
is not always unfavorable; a recent survey revealed a large number of examples of this
type of conversion”. The c.i. mass spectrum of apiitol pentaacetate is similar to those of
the common alditol acetates, in particular xylitol and arabinitol pentaacetates. The
interpretation of these spectra has been extensively discussed'°"7"°.

Separation of apiitol pentaacetate from xylitol pentaacetate by gas chromatog-
raphy is difficult, but complete or almost complete resolution of the two was achieved
with three columns. On the polar columns, SP-2330 and SP-2380, apiitol tetraacetate
overlapped other compounds because of tailing (Fig. 5). However, when acetylation
procedure 1V is used apiitol tetraacetate will not be present.

In previous analyses of natural materials apiose may not have been detected either
because the acetylation of apiitol was incomplete and the resulting apiitol tetraacetate
was not detected (or not identified), or because apiitol pentaacetate was not separated
from xylitol pentaacetate, or both. With the methods described here, the separation,
identification, and quantitative determination of apiose is possible.

ACKNOWLEDGMENTS

Supported in part by the Michigan Agricultural Experiment Station. Mass
spectral data were obtained at the Michigan State University Mass Spectrometry
Facility. The Facility is supported in part by grant DRR-00480 from the National
Institutes of Health. We thank Rawle I. Hollingsworth of this department for his
invaluable assistance in the identification of 3-0-(methylthiomethyl)apiitol tetraacetate
and Douglas A. Gage for performing the analyses on the gas chromatograph-mass
spectrometer. ‘

REFERENCES

l R. B. Duff, Biochem. J., 94 (1965) 768-772: .1. S. D. Bacon and M. U. Cheshire. Biochem. J.. 124 (1971)
555—562.

2 E. Beck, in F. A. Loewus and W. Tanner (Eds). Encrclopedia of Plant Physiology. Plant Carbohydrates I,

Vol. 13A. Springer-Verlag. Berlin. 1982. pp. 124-157.

M. W. Spellman, M. McNeil, A. G. Darvill. and P. Albersheim. Carbohydr. Res. 122 (1983) 115—129.

See, for example. G. Héfle and W. Steglich. Synthesis. (1972) 619-621.

P. Albersheim, D. J. Nevins, P. D. English. and A. Karr, C arbohydr. Res, 5 (1967) 340—345.

R. Wachowiak and K. A. Conners. Anal. C hem. 51 (1979) 27-30.

A. B. Blakeney. P. .1. Harris. R. J. Henry. and B. A. Stone. Carbohydr. Res, 113 (1983) 291—299.

D. L. Neal and P. K. Kindel. J. Bacteriol.. 101 (1970) 910-915.

W. Steglich and G. Héfle, Angew. C hem.. 81 (1969) 1001; Angew. Chem. Internal. Ed. Engl., 8 (1969) 98 l.

K. A. Conners and N. K. Pandit. Anal. Chem. 50(1978) 1542-1545.

.1. D. Albright and L. Goldman, J. Am. Chem. Soc. 87 (1965) 4214—4216: 89 (1967) 2416—2423.

—O¥DMQO\M#W

93

ACETYLATION OF APllTOL 65

12 A. J. Mancuso and D. Swern, Synthesis, (1981) 165—185.

13 P. M. Pojer and S. .1. Angyal, Aust. J. Chem, 31 (1978) 1031-1040.

14 K. Yamada, K. Kato, H. Nagase, and Y. Hirata. Tetrahedron Lett., (1976) 65-66.

15 R. Pummerer, Ben, 43 (1910) 1401—1412; L. Homer and P. Kaiser. Ann., 626 (1959) 19—25; S. an, T.
Kitao, S. Kawamura, and Y. Kitaoka. Tetrahedron, 19(1963) 817-820.

16 L. S. Golovkina. O. S. Chizhov, and N. F. Vul'fson. Bull. Acad. Sci. USSR. Chem. Sci, (1966)
1853-1863; Izv. Akad. Nauk SSSR. Ser. Khim., (1966) 1915-1926.

17 P.-E. Jansson and B. Lindberg, Carbohydr. Res, 86 (1980) 287—292.

18 M. Karni and A. Mandelbaum, Org. Mass Spectrom., 15 (1980) 53-64.

19 P.-E. Jansson, L. Kenne. H. Liedgren, B. Lindberg, and J. Lonngren, Chem. C ommun.. Univ. Stockholm,
No. 8 (1976) l—74a; Chem. Abstr., 87 (1977) 136153f.