:_-.-"-
:-:-;]1-
FOOD
HYDROCOLLOIDS
ELS tsVI E R
F o o d H y d r o c o l l o i d sl 5 ( 2 0 0 1 ) 4 4 1 . 4 5 2
www.el sevicr.crrn/locate/foodhytl
Structure and perfonnance of commercial kappa-2 carrageenanextracts
I. Structureanalysis
R. Falshaw^'*,H.J. Bixlerb,K. Johndrob
olndustrial
pO Box 3l-310, Lower Hutt, New Zealand
.blngretlients ResearchLtd.,
Solutions, Inc. (lSI), pO Box 407, Searsport,ME 01974, USA
Received7 December2000; revised l7 April 2001; acceptetll7 April 2001
Abstract
Extractsof two Chilean carrageenophytered seaweeds,Sarcothalia crispata and,
Gigartina skoîtsbergii,were prepare<Iby each of two
'industrialprocesses:
a mild alkaline extractionwith aqueoussodium hydroxide and a more vigorousuqu"ou. alkaline
extractionwith lime.
The resultingextracts'recoveredby precipitationwith isopropyl alcohol, were separatedinto gelling
and non-gelling fractions by leaching
with2'5o/aKCI' Theseprocesses
were also appiiedTochondrus crispusandto separatedgametophyteand sporophytesamplesof the Chilean
seaweedsfor comperativepurposes.The carrageenancompositionsof theseextractsand fractionswere determined
using both chemicaland
spectroscopictechniques ln addition, a set of decision rules is proposedand applieclto convert data from glycosyl linkage
analysisto
carrageenancomposition. The gametophyte extracts contained mixtures/hybridsof kappa and iota carrageenans(and
also mu an<l
nu carrageenans,dependingon the extraction conditions used). The tetrasporophyteextracts contained lambda
carrageenan(and also
theta carrageenan'dependingon the extraction conditions used). Extractive fractionationof mixed life phase
samples using 2.5vo KCI
'gelling'
yielded insoluble,
carrageenans
quite similar to thosefrom a gametophyteextractof the samespecies,but the soluble,non-gelling
fractions were not the sameas the correspondingsporophyteextracts.O 2001 Elsevier ScienceLtd. All rights reservecl.
Keywords:Cirrageenan;Kappa-2catrageenan;
Sarcothaliacrispota:Gigarrinaskottsbergii;Chondruscrlspas;Structureanalysis
1. Introduction
Industriallyimportantsulfatedgalactans,known as carra_
geenans,are found in numerousred seaweeds.
Traditionally,
Greek lettershave beenassignedto carrageenans
comprised
of certainidealisedcarrageenan
disacchariderepeatinguni$. I
A summaryof variouscarrageenan
structuresthat havebeen
assignedGreek letters was presentedby Craigie (1990).
However,nativecarrageenans
often containcombinationsof
theseidealisedunits,wiù variationsin structureoccurringnot
only betweendilïerentspeciesof seaweedbut alsowithin the
differentlife-stagesof a singlespecies.
Kappa, iota, mu and nu carrageenansare found in the
gametophyticlife phaseof variousseaweedspeciesin the
family Gigartinaceaeand the disacchariderepeat units of
thesecarrageenans
are shown in Fig. l. Mu and nu car-ra+ Correspondingauthor.Tel.: +64-4-569-06681fax: +64-4-569-005-5.
E-nuil utldress:[email protected] (R. Falshaw).
I
We entploy the Creek letterrlcsignationof itlealize<J
carrageenan
repear
u n i t s( R e e s ,| 9 7 2 )r a t h c rt h a nt l r en e w e ra n d r n o r ev e r s a t i l en o m e n c l a t u roef
K n u t s e n .M y s l a b o d s k i .L u r s e n .: r n d U s o r ' ( 1 9 9 4 ) . T h i s a r t i c l e i s a i m e d
largely at industrialresearchers
rvho lre rnore familiar with the Creek letter
n ( ) r n e n cl l l l u r c .
0 2 ( r t i ( X ) 5 X i o l / : i - s c c l . ( r n r 0 ) i L l l er i i l ( X ) l I l l s t : r i c r S c i c n c c [ - t t l . A l l r i u h t s
PII: S0l(rfi-005 X(0 I )00066-l
geenansare the biochemical precursorsof kappa and iota
carTageenans.
They both contain a sulfate ester group at
position-6 of a 4-linked d-D galactosyl unit which affects
the overall propertiesofthe carrageenanby creating 'kinks'
in the polymer chain that reduceits ability to gel. The tetrasporic life phase of Gigartinacean seaweeds.contains a
different type of carrageenan,most commonly lambda
cuurageenan(McCandless, Craigie, & Walter, 1973;
Pickmere,Parsons,& Bailey, 1973; Matulewicz,Ciancia,
Noseda,& Cerezo,1989).Lambda is the precursorof theta
carrageenan(Fig. l) but theta doesnot occur predominantly
in Gigartinaceanseaweeds.
6-Sulfated galactosyl units are converted to the conesponding3,6-anhydride
(3,6AG) using hot alkali. The rate
of conversion is dependentnot only on temperatureand
alkali concentrationbut also on the ionic strength of the
medium. Carrageenanstructure is also important, with the
conversionof lambda to theta being much slower than
the other conversionsshown in Fig. I (Ciancia,Noseda,
Matulewicz,& Cerezo,1993b).
The carrageenans
that occur in Gigartinaceanalgaeare
importantcommerciallyas theseparticularntixtttresor copolymers of kappa and iota carritgeenan(refèrredto as
,142
R. Falshaw ct ul. / Footl Hydrocolloùts t5 (2001) 441-452
alkali
-........-.---.--_-'
-20xh
Kappa (r<)
alkali
_____*_->
-50 x lq
Nu (v)
alkali
_....r-_--.>
k3
Lambda(I)
Theta(0)
Fig. I . Principal c:urageenansûuctures showing conversions that occur on alkaline treatment. Relative rates of conversion of mu to kappa and nu to iota are
given in terms of the rate constant, k3, for conversion of lambda to theta.
'kappa-2', 'kappa/iota
hybrid' or 'weak-gelling' kappa
carrageenans)have different properties to simple mixfures
of kappa and iota carrageenan from other sources. Long
reaction times and highly alkaline conditions are used
commercially to achieve a high level of conversion of
precursorsto kappaand iota crurageenanwhen gelling ability anùor protein reactivity of the carrageenanin foods are
being maximised. Milder conditions that produce less
conversionare used when cold solubility and viscosity are
the desiredresults.
In this paper,the structuralanalysisofcarrageenansfrom
three temperate-waterred carrageenophyteslSarcothalia
crispata (previouslyknown as Iridaea undulosa),Gigartina
skottsbergii and Chondrus crispusl is described. Several
studiescan be found on the structuralanalysisof the carrageenansfrom both hot and cold water extractsof individual
life phases of these species (Dolan & Rees, 1965;
McCandlesset al.,1913 ; Mafulewiczet al., I 989;Matulewicz,
Ciancia, Noseda,& Cerezo, 1990; Ciancia, Matulewicz,
Finch, & Cerezo, 1993a;Stortz & Cerezo, 1993; Stortz,
Bacon, Cherniak,& Cerezo, 1994). However, the carrageenansanalysedhere were producedusing commercial
life
extractionprocedureson mixed, as well as separated,
phase rnaterial,tbllowed by simple fractionation with
potassium chloride at one concentration to replicate an
industry test procedure.In the following paper, the strucfuresof theseextractsare related to some of their commercial
quality control and simulated dairy application properties.
2. Experimental procedures
2.1. Materials and methods
Two Chilean species, G. .skott.sbergii(known as
'pigskin')
and S. crispara (from Chiloe Island and known
'nama'
'sandpaper')were used. The third species,C.
as
or
crispus,was obtained from Western Nova Scotia,Canada.
The sampleswere obtained from bags or balesof commercially dried and cleaned,mixed life phaseseaweedstoredin
a Shembergplant in Cebu, Philippinesor in ISI's laboratory.
Different samplesof G. skottsbergii and of S. crispata were
used to produce the high viscosity (HV) and high milk
reactivity (HMR) processextracts.Dry weightswere determined by drying samplesat 105'C for 4 h. Clean weights
were determined by removing surf,acesand and salt and
extraneousmatter. Kappa carrageenanfiom Sigma Chemical Co. was usedas a referencematerial.
R. I;ulshau,et al. / Foul Ill,drocolloids t5 (2001) 441_452
2.2. E.ttructit>tt
The HMR processinvolved treatmenrwith lime t0.I M
Ca(OH):1,lor 30-48 h at 95 -t 2"C. The HV processuseda
lower concentration
of alkali (0.02M NaOH) but the ionic
strength was increasedby addition of sodium chloride
(0.l M). This process was carried out for 4-5 h at
95 )- 2"C. Samples of carrageenanobtained from these
extractionswithout any separationof seaweedlife phases
are denotedas 'mixed life phase'extractsand thosewithout
carrageenanfractionationas 'whole seaweed'extracts.
Threelevelsof extraction,differing primarily in batchsize,
were performed: (a) plant-scale using about 1000kg of
seaweed per batch (for HMR extraction of mixed life
phaseG. skottsbergiiand FfV extractionof mixed life phase
S. crispata),(b) laboratoryraw material testingscaleusing
about 100g seaweed(for HV extractionof mixed life phase
G. skottsbergiiand C. crispns, and HMR extraction of mixed
life phaseS.cispata awJC. crispus),and (c) bench-scaleusing
about 30 g of seaweed(for all extractionsof separategametophyte and tetrasporophyte
samples).
In all cases,the slurry of seaweedsolids and carrageenan
solution was filtered with filter aid in a pressurefilter. The
resulting clear but slightly tan coloured solution was
neutralisedwith HCI (4 M) to pH 8.0-8.5. In plant-scale
extractions, the carrageenanwas concentratedto about
2.5Vowlv by ultrafiltration but this step was omitted in the
smaller-scaleextractions.Carrageenanwas then coagulated
by adding pure isopropyl alcohol, the coagulumwas rinsed
with fresh alcohol, pressedto remove as much liquid as
possible,vacuumdried and powder milled (nominally 80
mesh).
2.3. Frac'tionation
In this work only a simpledivision of the carrageenan
into
gelling andnon-gellingfractionswas desired,so an extractive
procedureusing a relatively high concentrationof KCI was
powder( 10 g) was dispersedin a 2.5EoKCl
used.Carrageenan
(0.33 M) solutionat a concentrationof lVo w/v, and maintainedat 25"Cin a waterbathwith mild stining for l2 h. Perlite
filter aid was then addedand the mixture filtered.The filter
cakewas washedwith2.5o/oKCI solution.The carrageenan
in
the filtratewascoagulatedwith isopropanolandrecoveredasa
powderas describedabove.This carrageenan
was designated
the'non-gelling' fraction. The undissolvedcarrageenan
capturedin the filter ciùe was resiurriedin deionized(DI)
water.and heatedto 95"C. The hot solution was filtered,and
the filter cake again washedwith hot DI water. The carrageenanin the combinedfiltrateswas recoveredas aboveand
d e s i g n a t et d
h e ' g e l l i n gf r a c t i o n ' .
2.4. Seav,eed
s(porutiott
anclsporophytesof G. skottsbergiiand S.
Ciarrret<>çrhytcs
crisputa were separatedfrom dry commercial weeds by
visuiil insoectionof rewettcd fionds but this was not
44.1
possiblefor C. cri.sptts.Therefore,only the physically fractionatedgellingand non-gellingfractionsof extractsof this
speciescould be studied,and the quantityof the latterwas
usually too small for any applicationstesting.It is well
known that WesternNova Scotia populationsof C. crispus
contain very few sporophytes.
2.5. trC NMR spectroscopy
Spectrawere recorded on 3Vo w/v solutions in l:1 v/v
D2O:H2Oat 90'C on a Varian Unity-500 spectrometerat a
carbon frequencyof 125 MHz, using a 10 mm broad band
probe,acquisitiontime of 1.17s, delaytime of 0.8 s, and80"
pulse.Chemicalshifts are quoted relative to internalMe2SO
at 39.47 ppm.
2.6. Constituentsugar analysis
Sampleswere reductivelyhydrolysedwith N-methylmorpholine boranein aqueoustrifluoroacetic acid then acetylated to prepare alditol acetate derivatives as previously
reported (Stevenson& Furneaux, l99l), except that the
initial hydrolysisat 80'C was carriedout for 30 min to effect
an improvedrecoveryof the derivativesfrom 2-sulfated3,6anhydrogalactosyl(3,6AG) units (Jol, Neiss, Penninkhof,
Rudolph,& DeRuiter, 1999).The alditol acetatederivatives
were analysedby GLC using a Hewlett-Packard5890 Series
II chromatographwith flame ionisation detection and a
S u p e l c o S P - 2 3 3 0c o l u m n ( l 5 m x 0 . 2 5 m m ) a t 2 2 O " C .
Quantitation was based on experimentally determined
responsefactors(Falshaw& Furneaux, 1994).
2.7. Glycosyl linkage analysis
Sampleswere converted to triethylammonium salts and
methylated with MeVMe2SO-K- (Stevenson& Furneaux
l99l), purified by dialysis against H2Ox l; Et3NHCl
( 0 . 1M , p H 7 ) x 1 ; H 2 O X 2 a n d l y o p h i l i s e d .A s e c o n d
methylation and recovery was performed if, upon analysis
of the samples,it was determinedthat the first methylation
had been incomplete(by the presenceof galactitol hexacetate).The methylatedsampleswere reductively hydrolysed
and acetylatedas described above except that the initial
hydrolysistime was l0 min. The resultingpartiallymethylated alditol acetatederivatives were analysedby GLC as
above but at 170'C (l min) then 4'Clrnin to 220"C
(5min). To differentiateenantiomeric derivatives,C-1
deuteratedalditol acetateswere preparedfrom some methylated carrageenansamples by hydrolysis in TFA (2 M,
0.25ml, t h, 120"C) followed by reductionwith NaBDa
(15 mg/ml in aqueousI M NHIOH, 0.25 ml, I h, 25" C).
Afier the reductionwas quenchedwith acetone(0.5 ml)'
the samplewas evaporatedto drynessand acetylatedby
the method of Falshaw and Furneaux (1995), exccpt
that the organic layer was washed with watcr (4 rll),
NarCOr (0.5M. 4 rnl) and water (4 rnl). The partially
methylatedalditol acctate derivativeswere analysedby
+++
R. l"alslutyt'tr ul. / F'rntl llvdrocolloùls lS (2001)141-152
GLC, as above, and wcrc a l s o a n a l y s e d b y G L C - M S
( F a l s h a w& F u r n e a u x . 1 9 9 4 ) .
3. Results and discussion
-3.1
. Santpleeilra(îio,t
The yields of carrageenanobtained from HMR an<tHV
extractionsof gametophytic,sporophyticand mixed life
phase seaweedare shown in Table l, and are typical of
commercialas-isyields.As-is yields are basedon seaweed
weight including moisture, sand, salt and foreign matter.
The seaweedsunder investigation were from high quality
suppliers with moisture about 22o/owlw and the other
inrpurities5o/ow/w or less.
-1.2.Sample .frocrionati on
The weight percentof the gelling and non-gellingfractions obtainedusing2.5VoKCI from a sampleof eachmixed
life phaseextractand the total recovery relative to the starting sample are tabulated in Table l. Material balance
closure rangedfrom 96 to 1027o.
3.3. Chemicalanalysis
3.-1.l. Constinrcfi sltgar analysis
3,6-Anhydrogalactose
and galactosewere the predominant sugarsderived from all the samples.Small amounts of
xylose(l7o) andglucose(<5Vo)werealsoobservedin some
samples.More xylose(4Vo)wasobservedin the non-gelling
fiaction of the C. crispus HV extract and larger amountsof
glucose(9 and l37o) in the non-gellingfractions of the
HMR extractsof tetrasporicand mixed life phaseG. skottsbergii, respectively.The xylose is most likely to occur as
single brancheson the 3-linked galactosylunits in carrageenans.The glucoseis likely to be derived from floridean
starch present in these samples.In order to compare the
carrageenancompositionsof the various fïactions,the constltuentsugarresultswere normalisedto excludeglucose.
The resulting 3,6AG contentsof all the extractsand fractions analysedare shownin Table l.
The level of 3,6AG was close to 50Vofor the gelling
fractions of mixed lif'e phase extracts of all three species
extractedusing the HMR process (and also the whole
gametophyte
extracts,wherestudied).This is to be expected
fbr seaweedsthat contain kappa. iota, mu and nu carrageenans where the mu and nu cârrageenanshave been
convertedto kappaand iota, respectively,during the long
hot alkali conditionsof the HMR process.The ratios of
3,6AG to galactosewere less than 1:1 for the equivalent
sanrplesextractedusingthe HV process.The milder conditions usedhereevidentlydo not convertall the mu and nu
presentin the original seaweedto kappa and
cafrageenans
i t r t i tc u l r i r q c c n r r nr cs s. p ù c t i v c l y .
Thc wholc tetrlsporophyteextrâctsobtainedusing the
HV processcontainedlittle or no 3.6AG, as expectedlbr
lambda carrageenan.The whole tetrasporophyteHMR
extractsdid containsignificant^mountsof 3,6AG but not
closeto Ihe50Voexpectedif all the lambdacarrageenan
had
beenconvertedto thetacarrageenan.
This reflectsthe slower
rate of conversionof lambda to theta (Fig. l). Basedon
these values, the non-gelling fractions (for both the HV
and HMR processes)
of all the samplesanalysedcontained
more 3,6AG than would be expectedif they containedjust
lambda and some theta carrageenans.The percentgelling
fractions obtained for both HMR and HV extracts of C.
crispus were higher than for G. skottsbergii which, in
turn, were higher than for S. crispata. This progression
would be expectedfrom the 3,6AG levels in thesesamples
(as well as from the relative performanceof the extractsin
the applicationstests,see following paper).
The relatively large changein the amountof gelling fraction obtainedfor S. crispata in going from an HV extract tcr
an HMR extract,comparedwith the smaller changefor G.
skottsbergiiand the even smallerchangefor C. crispas,does
not seemto coincidewith the changesin 3,6AG content.For
instance,it would be expectedfrom the progressionsin the
yield of the gelling fractions that the 3,6AG increase in
going from HV to HMR sample would be highest for S.
crispata and lowest for C. crispus. However, from the
results in Table l, the 3,6AG changes are roughly the
samefor all three seaweeds(10-14 moleVo).Most likely,
this behaviour is related to where along the chains with
gelling potential the 'dekinking' by 3,6AG formation took
place.That is, a relatively small amountof 3,6AG formation
could have resulted in the formation of large blocks of
kappa and iota carrageenanson alkaline treatment of ,S.
crispata; whereas for C. crispas large blocks of gelling
structuresevidently exist before alkaline treatmentand are
not much changedby the treatment.
3.3.2. Glycosyl linkage analysis
In the sampleswhere glucosewas observedby constituent sugaranalysis,the partially methylatedalditol acetate
species4-Glc was observed,as expectedif florideanstarch
were present.The data shown in Tables2 and3 was normalised to exclude4-Glc in order to allow direct comparisonof
the carrageenancompositionsof all the samples.The data
for whole gametophyte and tetrasporophyteextracts are
consistentwith thoseobtainedpreviouslyfor thesespecies
(allowingfor differencesin sampleorigin,extractionconditions and analyticalmethods)(McCandlesset al., 1973;
Matulewiczet al., 1990; Ciancia et al., 1993a).The total
3,6AG contentswere closeto thoseobtainedby constituent
sugaranalysisfor HMR extractsbut not for HV extractsas
,1-linkedgalactosylunits
additionalconversionof 6-sulf'ated
to the corresponding3,6-anhyclridesoccurred in these
sarnplesduring methylation.This was confinned by the
in all
absenceof 4,6-Gal (indicativetlf rnu carrageenan)
t h e s es a m p l e sH. o w e v e r .s o m e2 , 4 . 6 - G a(l i n d i c a t i v eo l n u
was observed.This rellccts the
or lambda can'ageenuns)
R. I;ulshLrwet al. / F-oodH y*dnttolloitlsI 5 (2001) 1l t *152
o
o
o
: 9 È
-t .S
F i ,
8 3 3 3 r3 Ê8
i
l
t
i
t
l
t
,F
l
6e
€
€o
U
Ê
d
< t r
di5
o
' Ë € + * æ É
È
I
I
I
I
I
I
I
I
\ o o c - æ Ê o
o n
r
|
|
I
I
I
|
|
ç
Ê
Ê
N
4
!
tr
-
o
è0
E
z
È
> ù
U
!
)
6R
o
I
F
!
o
Ô
o
o
o
<tr
Ô
J
t \ æ æ r r s
$ $ . + ô o -
|
|
|
I
I
|
|
{
I
U
€
,N
-' é >È
a
N
!
o
(]
>:)
o \ æ r o . r n
|
|
|
|
|
|
|
|
.
@
o
ô
èa
!
€o
!
i
./)
8
o
-É
'AE
.9
<tr
ri3
E
o
o
\ N N \ N q O O . o N h m È C ,
Y Y Y h ' t N n $ * ' - o ô
-
;
a
q
"
x9
o
9
a
U
>:-
N
N
d
o
N
S
d
æ
È
i
Q
o
N
x
o
o v s ô r d o o s o ô o o o
c
ù
o
o
e
e
e
è
p
p
-
û
'ii
o
!
p
U
!
> *
b È e
a
o
o
o
à > o o > , > , o o
i { â > >::
> >
d
o
* À
o
X
X
X
X
X
X
É
E
H
H
E
E
h
h
o
a
-
o
+
0 )
x
= i - l
4
€
x
È
È
h
a
L a ^ - 9
o
\
I
C
O
U
c
ô.9E
o
o
!
9
a
6
9
.
€
8
! >
F
N
O
.
{
€
r
ù J
- = \
ù
\
^
'
ù
' \ i
I
ù *
L ' v i
ù
!
-
ù q
{ e
l
-
L
ù
ù g
_ \
M H
ù
M H
È
€ € 3 È
> D
a e
à
: i* fà +Èt >
Èi + * iÈ- ! * i È
*È È& *<6 > - c Ao
\ {
{
{
U
q
a
t - r O . a U O - O O O . ; U . ; O r ;
:l;16
R. F'ulsln-- et ul. / I.'uxl Hr-drot:olloitlslS (2001)
14t_452
Tablcl
Glycosytlirrkageanalysisclfwholegarnetophyte(G)extractsandgellingfractions(Gel)ofmixedlrfephase(Mixc<l)samplesofC'.r.ri.r2rrr(Cc
(Gs) and s' cilsputo (Sc) Numbcrs in bold
relate to the major idealisedcarrageenanstructures
Normalised moleo/o
Species
Lifè Phase
Extrâction
Fr:rction
Constitucntsugar and
d e d u c e ds u b s t i t u t i o n "
2,3-Gal
3,4-Gal
3-Cal
3,4,6-Galb
2,3,6-Gal'
Total 3-linkcd units
,l-AnGal
2,4-AnGal
2,4,6-Gal'
4,6-Gal
4-Gal
2,1-Gal
'l'otal
4-linked units
2,3,4-cal
2,3,4,6-Gal
Total ambiguous linkage units
Carrageenan
v
F,a,E,y
z, pyruvated o
À -type
K , ( p ' r ) ,p
t, 0, (u"),a
À,u,y
p,ô
ç
Cc
Mixed
HV
Cel
Gs
Mixed
HMR
Gel
cs
Mixed
HV
Gel
Gs
U
HMR
Whole
Cs
G
HV
Whole
I
46
3
0
o
50
38
t2
0
0
0
0
50
0
0
0
I
47
3
0
o
5I
38
9
2
0
0
0
49
0
0
0
2
39
1
45
3
0
0
49
35
10
6
0
0
0
51
0
0
0
0
46
4
0
0
50
33
16
0
0
0
0
49
l
0
1
4 4
4
t
0
49
34
10
5
0
0
0
49
l
lr
2
Sc
Mixed
HMR
Gel
type
À ,0
K' L, p,
Cc
Mixed
HMR
Gel
4
0
0
36
17
I
0
0
0
54
I
0
I
o
2
4
9
l
0
0
52
29
t7
l
0
0
0
47
t
0
1
Sc
Mixed
HV
Gel
Sc
C
HMR
Whole
Sc
G
}IV
Whole
0
43
o
I
46
2
o
0
49
32
t7
0
0
0
0
49
2
0
2
0
47
2
I
I
45
35
ll
5
0
I
I
53
I
lr
2
I
0
50
33
ll
5
0
0
0
49
I
0
I
a 2,4-disubstituted
and/orlinkedgalactopyranosyl
unir,anatysed
as 1,2,4,5-tetra-o-acetyl-3,6-di-o
*"*rtg"*,u"*"
, f i^.!:lf":
Assumed
to bepyruvared
[4,6-o-(carboxyethyridine)]
3-rinkedgalactosyl
unit.
'
Enantiomericpartially methylatedalditol acetates(2,3,6-Galand 2,4,6-Gal)
differcntiâtedby deuteriumlabelling and determinedby GC-MS analysrs.
'
Lonverted to K-carageen:lnduring processingor glycosyl linkage analysis.
'
Convertedto r-carrageenanduring processingor glycosyl linkage analysis.
r
A secondmethylationof thesesampresresurtedin degradationof the sample.
slower conversionof lambda to theta and nu to iota than mu
to kappa carrageenan.
Most notably, the compositionsof the non-gelling frac_
tions of the mixed life phase HV extracts were distinctly
different from thoseof the correspondingwhole retrasporo_
phyte extracts, with significant amounts of 3,4_Gal(and.
thus, kappa,iota, mu and/or nu carïageenans)present.
3.3.3. Data analysis
3.3.3.1.Rulesfor determining extract composition. Most
of the derivativesobservedin Tables 2 and 3 can be attributed to more than one of the idealisedc,urageenandisacchariderepeatunitsshownin Fig. l. For example,Z,4,6-Gal
can be derivedfrom both lambdaand nu carrageenans
since
lambda: 2,3-Gal+ 2,4,6-Gal and nu :3,4-Gal + 2,4,6Gal. Obviously,therecould be just lambdaor just nu or a
c o m b i n a t i o no f b o t h c a r r a g e e n a npsr e s e n ti n a n y g i v c n
sampledependingon the relative amountsof 2,3-Galand
3,4-Galpresent.Numeroussimultaneousequationscan be
constructeddepending on the number and type of carrageenanstructuresdefined.
All the samples examined here also contain some
miscellaneous3-linked or 4-linked monosaccharide
units
thatdo not con'espond
to any of thoseshownin the idealised
carragcenan
structures
of Fig. L Someof the units,suchas
3-Gal, can be attributed to a number of other, known
carrageenanstructures(such as alpha, beta, gamma or psicilrageenans) (Craigie, 1990), while other units may be
componentsof carrageenanswith idealised structuresthat
have not been assignedGreek letters.In addition, there are
units with ambiguouslinkage/substitutionpatterns.2,3,4Gal, for example,is derived from a galactopyranosylunit
that is either linked or substitutedat rhe 2-. 3- and 4positions.It could correspondto a 2,4-disulfated3-linked
galactopyranosylunit or, a 2,3-disulfated4-linked galactopyranosyl unit as observedin the polysaccharidefrom the
red seaweed,Champia novae-zealandiae(Miller, Falshaw,
& Furneaux, 1996). Since seaweedsare natural organisms
there is always potential for structuresthat do not fit in a
rigid naming system.These unusualmonomersmay not be
containedin homogeneouspolymers but scatteredalong a
chain of other carrageenanstructuresand act as spacers
where they may interrupt double helix formation, and thus
prevent,or weakengelation (Lawson, Rees,Stancioff,&
S t a n l e y1
, 9 7 3 ;P e n m a n& R e e s ,1 9 7 3 ) .
The estimationof the relativecompositionsof complex
polysaccharide
mixturescan be conductedusing setsof predefineddecisionrules [e.9. cell wall compositionof grape
berries(Nunan,Sims, Bacic, Robinson,& Fincher,1998)1.
Cianciaet al. (1993a)calculatedthe contributionof various
carrageenan
tlisacchariderepeatunitsto of threegametophyte
R. f ul.shawet ul. / f'ood llyt!rut'olloitl.st5 (2001) 11t-452
Table 3
Glycosyl linkageanalysisof whole tetrasporophyte(T) extractsand non-gellingfractions(Non-Cel)of mixed life phase(Mixed) sarnplesof
(Cc). C
C', cri.ypus
skott.rbergii(Gs) and S. crispttltt(Sc). Numbers in bold relate to the majoridealised carrageenansrructures
NormalisedmoleTo
Species
l-ifè Phase
Extraction
Fraction
Constituentsugarand deduced
substitution'
2,3-Gal
3,4-Gal
2,3,6-Galb
3-Gal
Total 3-linked units
4-AnGal
2,4-AnGal
2,4,6-Galb
4,6-Gal
4-Gal
2.4-G.,n
Total 4-linked units
2,3,4-Ga1
2,6-Gal
Terminal Xyl
Total ambiguous linkage units
Cc
Mixed
HV
Non-Gel
Gs
Mixed
HMR
Non-Gel
Gs
Mixed
HV
Non-Gel
Gs
T
HMR
Whole
Cs
T
}{V
Whole
Sc
Mixed
HMR
Non-Gel
Sc
Mixed
HV
Non-Gel
Sc
T
HMR
Whole
Sc
T
HV
Whole
19
l8
42
3
3
10
6
0
4
4
0
M
2
3
2
47
2
3
4
0
0
0
48
3
2
0
5
2
3
l
50
3
1
8
3
U
0
0
0
45
2
3
0
5
25
t4
42
0
39
0
5
5
0
47
1
1
37
0
0
0
48
2
3
0
5
l
45
0
3
48
0
0
0
51
2
2
0
4
Carrageenan type
À,0
K, L, lL, V
À+ype
B,a.6,y
r , ( p ' ) ,É
t. 0, (vd), a
A, U , y
p,ô
E
)
ll
50
9
6
23
0
3
2
43
6
o
I
7
3
I
J
49
5
1
28
0
0
0
47
4
o
0
4
49
5
5
2l
0
0
0
5I
0
0
0
0
2
4
8
0
2
o
50
2
1
5
2
9
0
0
0
46
4
o
0
4
/
46
l5
4
29
I
I
0
50
2
z
0
4
0
' 2,4-Gal means 2,4-disubstituted
a
and/or linked galactopyranosyl unit, analysed as 1,2,4,5-teu:a-O-acetyl-3,6-di-O-merhyl-gâlactitoletc.
b Enantiomeric partially
methylated alditol acetates(2,3,6-Gal and 2,4,6-Gal) differentiated by deuterium labelling and determined by GC-MS analysrs.
"
Converted to K-carrageenanduring processing or glycosyl linkage analysis.
'r
Converted to i + ca-rrageenan
during processing or glycosyl linkage analysis.
fractions of G. skonsbergiibut the processof calculation was
not described.Vy'eproposehere a set of decision rules, by
which the proportions of various combinations of carrageenansunderstudyherecanbe estimated,basedon idealised
carrageenanstructures.While the derived conclusionssuffer
from certainsystematicproblems(as explainedbelow), they
can certainly be usedfor comparalive purposessince they are
calculatedin a consistentmanner.The followins rules were
appliedto the datain Table 2:
minus the percentage of lambda and nu ciurageenans
alreadycalculatedfrom Rules (1) and (3) (as thesedo not
contain3.6AG either);
6. the total moleVoof ambiguous/miscellaneouscilrageenans
was the mole%oof 3-Gal + 3,4,6-Gal+ 2,3,6-Galx 2 (as
thiswasalwayslargerthanthe mole%o
of 4-Gal + 2,4-Gal),
or thetotalambiguousunits x 2, whicheverwasthegreater;
7. themole%okappacarrageenanwas 1007ominus Vo\ambda,
Vothera.Vonu Voiota. Vomisc. and Vomu, tf 3,4-Gal > 0.
L the moleTo lambda carrageenan was the moleVo of
2,4,6-Gal(to the level of 2,3-Gal),plus an equal moleTo
of 2,3-Gal;
2. the moleVotheta carrageenanwas the moleTo of 2,3-Gal
remaining unassignedafter applicationof Rule l, up to
thelevel of 2,4-AnGal,plus an equalmoleToof 2,4-AnGal;
3. the moleTonu carrageenanwas the moleToof 2,4,6-Gal
remainingunassignedafter applicationof Rule l, up to
the level of 3,4-Ga1,plus an equal moleToof 3,4-Gal;
was the moleToof 2,4-AnGal
4. the moleVoiota carrageenan
(remainingunassigned
application
of Rule 2) up to the
after
afterapplication
unassigned
that
remained
level of 3,4-Gal
of 3,4-Gal;
of Rule 3, plus an equal moleo/o
5. the moleo/omu carrageenanwas twice the difference
between iclealised3,6AG moleo/o(50tk) anclthe actual
3,6AG moleToby constituentsugar analysis(tf <48o/o,
basedon the valueobtainedfbr Sigma kappacarageenan)
This methodtendsto underestimatetheVo nu, due to alkalimodification during methylation and thus overestimatethe o/o
iota. If the Vonu is underestimated,the 7o mu will be overestimatedbecauseof Rule 5. As an internalcheck,the moleo/o
of 4-AnGal x 2, should equal the total Vokappa + o/omu (as
mu is convertedto kappa due to alkali-modificationduring
methylation).In most sarnplesthis was the case.
3.3.3.2.Gametophyteextracts and gelling fractions. Ustng
theserules, 100Voof the units could be assignedin all the
samples,Table 4. As expected, the whole gametophyte
A
HMR extractscontainedkappa and iota carrageenans.
small amount of theta carrageenanwas also assignedto
the whole gametophyteHMR extract of S. cri'spata.The
gelling fractions of mixed life phase samplcs from
HMR extractswere similar to the corlespondingwholc
gametophyteHMR extracts,except that small atnountsof
l.llt
R. F'ulslutwct ul. / t'ood L!1,tlrot:olloid.s
tS (2001) ,11t*:.52
Table J
Idealiscd canageenân structures in gelling fractions and gametophyle
extracts
Seaweed
species
C. crispus
C. skottsbergii
C. skottsbcrgii
S. crispata
S. crispata
C. trispus
G. skottsbergii
G. skottsbergii
S. crispata
S. t'rispala
Seaweed life
phase
Extraction
Extractive
process
fiaction
Mixed
Mixed
Carnetophyte
Mixed
Gametophyte
Mixed
Mixed
Gamerophyte
Mixed
Gametophyre
HMR
HMR
HMR
HMR
HMR
HV
HV
HV
HV
HV
Gelling
Gelling
Whole
Gelling
Whole
Gelling
Gelling
Whole
Gelling
Whole
lambda
(molea/o)
theta
(mole%o)
0
2
0
2
0
2
2
0
2
2
0
0
0
0
0
z
2
0
0
0
theoretically non-gelling lambda./thetacanageenan were
also present. Either this lambda/thetacarrageenanexists
in somekind of gellingmoleculeor somenon_gelling
mate_
rial was retainedin the gelling fractions. Similar observations have been made previously (Stortz & Cerezo, 1993).
One consequenceof setting defined rules is the creationof
anomalies.Thus, becausethe observedmoleVo3,6AG was
slightly lower rhanexpeÇted,the dataanalysisindicatesthat
the gelling fiaction of the HMR exrractof mixed life phase
C. crispus contains 67o mu ciurageenan.This, in fact, is
unlikely, given the conditions used to extract the sample.
As expected, the whole gametophyte HV extracts
containedkappa, iota, mu and nu carrageenans.
Again, the
gelling fractions of mixed life phase samples from HV
extractsof the th_ree
specieswere similar to the corresponding whole gametophyte HV extracts, except that small
amounts of theoretically non-gelling lambda carrageenan
were sometimespresent.
3.3.3.3. Tet r asporophyt e ext rac ts and non-g ellin g frac tions.
The resultsof the data analysisare shown in Table 5. For
these samples,lambda was calculatedas the mole%oof
2,4,6-Galplus an equal amount of (2,3-Gal+ 2,3,6-Gal),
if these were >0, in order to minimise the level of
miscellaneouscarrageenans.
This is reasonablebecause
lambda carrageenans
from several speciesin the Gigartinaceae are known to contain 2,6-disulfated3-linked
units (Matulewicz et al., 1990; Stevenson& Furneaux,
l99l; Falshaw & Furneaux, 1994). A proportion of the
3-Gal present should also be assigned to 'lambda'
carrageenansince unsulfated 3-linked units are known
to occur in the 'lambda' carrageenanfrom C. crispus
(l)olan & Rees. 1965).However, without data on equivalent samples fiom gametophyte and tetrasporophyte
extracts fiom C. crispus, it is impossible to calculate
how much 3-Gal to assign to 'lambda' carrageenan.
Therefore,it has all been consideredas 'miscellaneous'
even though this results in a large amount of miscellangousunits for the non-gellingfiaction of the C. crispus
H V e x t r a c t . I n s o m e i n s t a n c e s ,t h e t o t a l c a l c u l a t e d
c a r r a g e e n â ne,v e n i n c l u d i n gr n i s c e l l a r r e o u
s rrageenans,
ca
nu
rota
(moleo/o) (ntoleTc)
0
0
0
0
0
1
1
1
1
2
3
3
3
3
2
0
0
0
0
2
2
2
2
2
2
2
2
2
18
0
0
2
2
1
2
2
2
Inu
(moleTo)
kappa
(rnoIe7o)
ntt sc.
(mole%,)
Total
(rnoieT')
6
0
0
0
0
22
4
0
2
4
64
56
60
62
62
50
48
40
38
38
6
8
IJ
2
4
6
6
l0
8
6
I00
100
r00
100
100
100
100
100
100
I00
does not total l00%o.This highlights another deliciency
of applying rigid rules ro such data.
As expected,the two whole tetrasporophyteHV extracts
containedpredominantlylambda carrageenan.Someof this
lambda carageenanwas convertedto theta ciurageenanin
the correspondingwhole tetrasporophyteHMR extracts.
The non-gelling fractions from the HMR extracts of
mixed life phaseG. skottsbergiiandS. crispata weresimilar
to the correspondingwhole tetrasporophyteHMR extracts
but contained small amounts of mu and/or kappa carrageenan.(There was insufficient non-gelling fraction from
the HMR extract of mixed life phase C. crispus to be
analysed.)Most notably, all the non-gelling fractions from
the HV-processedmixed life phase seaweedscontained
significantamountsof kappa, iota mu and nu carrageenans.
The most plausibleexplanationfor this behaviouris that the
mu, nu, kappa,and iota hybrid polymers found in this fraction are less 'blocky', i.e. more randomlydistributedand,
thereby,more soluble in 2.5o/aKCl.
Whatever the explanation,it is clear that the extractive fractionation procedure used here does not produce
a non-gelling fraction identical in composition to the
carrageenan present in the sporophyte of the same
weed. It does, however, yield a fraction whose structural analysis is of industrial significance and whose
makeup is important in interpretingcertain applications
data (see following paper).
-1.3.3.1.Total extracts.The data in Tables 1,4 and 5 were
used to estimate the carrageenancompositions of total
extractsfrom mixed life phaseseaweeds,Table 6. For the
HMR extractions,the compositionsof G. skottsbergiiandS.
crispata carrâgeenans
are quite similar. On this basis,they
would be expectedto perform similarly in product applications. The HV-extractedS. crispata carrageenancontained
much lesskappa,and much more lambdathan the equivalent HV-extractedG. skottsbergiicarrâgeenanor the HMRextractedS. crispata. This is most likely due to different
proportions of gametopbytic and tetrasporic seaweed
presentin the variousraw materialsused.The perfilrmance
of theseextractswould be exoectedto bc dill'crent.The
R. F'ul:ltaw et ul. / Food Hy-tlrotolloitl.çl5 (200t) 14t_452
Table5
Idcaliscdcarraget:nanstructuresin non-gelling fractionsand tetrasporophyteextracts(nd:
not detennined)
Searvecd
spccies
Seaweedlitè
phase
Extr;rction
process
Extractive
fraction
larnbda
(moleTo)
theta
(mole%)
nu
lota
mu
kappa
misc.
Total
(moleo/o) (moleTc) (moleVo\ (moleô/c) (rnole%) (mole7o)
Ç. cri-spus
Mixed
Mixed
Sporophyte
Mixed
Sporophyte
Mixed
Mixed
Sporophyte
Mixed
Sporophyte
HMR
HMR
HMR
HMR
HMR
HV
HV
HV
HV
HV
Non-gelling
Non-gelling
Whole
Non-gelling
Whole
Non-gelling
Non-gelling
Whole
Non-gelling
Whole
nd
56
58
48
14
42
20
86
)o
88
nd
28
30
20
0
0
0
0
0
nd
0
0
0
0
4
22
a
z
0
G. skottsbergii
G. skottsbergii
S. crispata
S. crispala
C. cri:-plrs
G- skottsbergii
G. skottsbtrgii
S. cri,spata
S. crispato
-tt
t
nd
0
0
0
0
2
l0
4
8
0
nd
6
0
0
0
'1
nd
2
0
6
0
z
0
6
0
40
0
t8
0
I J
nd
8
8
l0
l0
22
o
l0
lo
8
nd
100
96
r00
104
100
100
100
100
96
Table 6
Estimated idealised carrageenan structures in whole seaweed extracts
Seaweed
Seaweed life
specles
phase
Extraction
process
Exûactive
fraction
lambda
(rnole7a\
theta
(mole7o)
nu
(mol.-7o)
iota
mu
kappa
misc.
Total
(moleVo) (mole%o) (moleo/o) (mo\e%o) (mole7.a)
C. cri.spus"
G. skottsbergii
S. t:rispata
C. crispus
G. skottsbergii
Mixed
Mixed
Mixed
Mixed
Mixed
Mixed
HMR
HMR
HMR
HV
HV
HV
Whole
Whole
Whole
Whole
Whole
Whole
0
9
l0
5
8
28
2
5
8
0
0
0
0
0
0
2
t4
o
22
26
25
l7
6
I
0
20
1"7
l 1
t5
t4
'
64
46
50
46
41
28
o
8
3
7
o
100
96
96
98
102
100
Basedon gelling fraction only
consequencesof the different compositions of these
materialsare discussedin the following paper.
3.4.
rrC
NMR spectroscopy
Ciancia et al. (1993a) tentatively assignedminor rrC
NMR signalsin carrageenanfrom gametophyticG. skottsbergii to a number of carrageenanstructures(most without
designatedGreek letters),based on chemical shifts calculated from a model of idealiseddisaccharideunits (Stortz &
Cerezo, 1992). However, more recent, experimentallyderived resultsindicatedeficienciesin this model (Falshaw
& Furneaux, 1994; Stortz et aI., 1994; Falshaw,Furneaux,
Wong, Bacic,Liao, & Chandrkrachang,
1996).The limit of
detectionis now generallyconsideredto be around 57o,so
minor componentsare often undetected.
Also, lambda-type
carrageenanscan be virtually invisible in mixtures with
other types of polysaccharides,due to unfavourable spin
relaxationpropertiesof highly viscoussamples.However,
''C
NMR spectroscopyprovides the opportunity to independentlycheckthe majorcarrageenan
compositionscalculated above.
-).1.1.Gelling.f'roctionsand gametophyteertracts
The whole gametophyteextrâctsand gelling fractionsof
mixed lit'e phase samples of the. three species extracted
usingtheHMR processproduced C NMR spectrashowing
signalscorrcspondinsto thoseexpectedfor kappaand iota
cârrageenan(LIsOv & Shashkov. I 98-5).There was no
evidence of mu/nu carrageenanin the spectrum of the
HMR-extracted mixed life phase C. crispus, suggesting
that the mu carrageenanin Table 6 is an anomaly resulting
from applicationof the rules set out in Section3.3.3.1.The
equivalentHV-processedsamplescontained signalsconespondingto thoseexpectedfor kappa,iota and mu/nu carrageenans(Cianciaet al., 1993a).Mu and nu czurageenans
are
not resolvedusing this technique.The lambda and/or theta
carrageenans
presentin some of these sampleswere at too
low a level to be detected.
3.4.2. Non-gellingfractions and tetrasporophyteextracts
Spectraof the whole tetrasporophyteHV extractsof G.
skott.rbergii and S. crispata were both poorly resolved but
showed signals characteristic of lambda carrageenan
(Falshaw & Furneaux, 1994). Surprisingly. a signal at
72.0 ppm was observedin the spectrumof the whole tetrasporophyteHV extract of G. skottsbergii. This signal is
chtuacteristic of iota carrageenan (Usov & Shashkov,
1985) but would not be expectedin a samplefrom only
tetrasporic
seaweed. However,
the calculated
carrageenan
compositiondoes include a small amount of iota carrageenan.The origin of this iota carrageenanis unclear but
could have arisen from less than perfect separation of
gametophyticand tetrâsporicplants. Spectraof the whole
tetrasporophyte
HMR extractsof G. skottçbergiiandS. cri.çpata were also poorly resolved,but showedsignalscharacteristic of lambda and theta carrageenan,as cxpectcd
(Falshaw& Fumeaux. 1994).The spectrumfor thc whole
R. l-al"-huw ct al. / F-ooclHydrocolloids I 5 (200I ) .t4 I ..452
S. cispata non-gellingfractionof HV elitract.
trft-G1
r-G1
x-G1
x-41
p/v-D1
r-41
l.-D1
G. skotfsbergl non-gellingfraction of HV extract.
1lv-G1
Iil
il
r-c1
/\I
rlf,ff^rf"*./
*
105
100
90
85
f0
55
ppr
ltCNMRSpectra(50 ll0ppmregion)ofG.skottsbergiia;nds.crispatanon-gellingfractionsofHVextracts.Keysignalsarelabelled(G:3-linked
Fig.2.
p - o - g a l a c t o p y r a n o s yDl .: 4 - l i n k e d a - D - g a l a c t o p y r a n o suynl i t s ,A : 3 , 6 A G u n i t sa n d F S : f l o r i d e a ns t a r c h ) .
R. L'ulshuw,et ul. / f ood llytlroco!toitl.s l5 (2()01) 411_152
tetrasporophyte HMR extract of G. skottsbergii also
containedsignalscharacteristic
of florideanstarch(Falshaw,
Furneaux,& Stevenson,
1998),asexpectedfrom constituent
sugaranalysisof this sample.
Spectraof the non-gellingfractions of HMR extracts
from mixed lif'e phaseG. skottsbergiiand S. crispata were
reasonablywell resolvedand showed signals characteristic
of lambdaand thetacarrageenan,
as expected.However, the
resolutionwas insufficientto conclusivelyassignsignalsto
the C-l of 4-linkedunits in kappacarrageenan
(95.3ppm)
as opposedto theta carrageenan(95.6 ppm) or iota carra_
geenan(91.9ppm) cf. lambdacarrageenan
(91.7ppm). The
spectrumof the non-gelling fraction of HMR extractedG.
skot tsberg i i alsocontainedsignalscharacteristicof fl oridean
starch,as expectedfrom constituentsugar analysis.The
spectrumof the non-gelling fraction of HMR extracted.S.
crispata containedan unassignedsignal at 99.9ppm. This
was distinct from the C-l of 3-linked units in theta carra_
geenan(100.3ppm). A signal at99.g ppm is characteristic
of the C- I of 4-linked B-o-glucopyranoseunits in floridean
starch but no glucose was observedby constituent sugar
analysisof this sample.
Spectraof the non-gelling fractions of HV extractsfrom
mixed life phase G. skottsbergii and S. crispata were
reasonablywell resolved,but complex (Fig. 2). Signals
characteristicof lambda, kappa, iota, and mu/nu carrageenanswere all observedbut not those for theta carageenanin both cases.This is, in fact, consistentwith the
calculatedcarrageenancompositions.A signal at 99.9 ppm
characteristicof floridean starch was also observedin the
spectrum of non-gelling HV extracted G. skottsbergii.
Glucose (2 moleo/o)was observed by constituent sugar
analysisof this sample.There was insufficientnon-gelling
HY C. crispu.sextract on which to record a '3C NMR spectrum of this material.
4. Conclusions
A range of carrageenansampleshas been prepared by
extractionusing typical commercialprocessesand subjected
to structuralanalysis.It was shown that the extractivefractionationof mixed life phasesampleslusing2.57oKCI yields
gelling carrageenans
quite similar to those from a gametophyteextractof the samespecies.However,the non-gelling
fractions are not the sameas the correspondingsporophyte
extracts.In general,calculatedcarrageenancompositions
'rC
were consistentwith thoseevidentfrom
NMR spectroscopicexarninationof the conespondingsamples.
Acknowledgements
Chemicaland spectroscopic
analyseswere conductedat
IndustrialResearchLtd., Lower Hutt, New Zealandunder
contractto Ingreclients
Solutions.Inc. The authorswould
like to acknowledgethe extractionspelfbnned under the
4,5I
direction of Engineer Robina Navarro in the Shemberg
BiotechCorp. Carmenfacility. We also want to acknowl_
edge the laboratory assistance of Birgitta Oclom and
Shannon Barber from ISI, and the NMR spectroscopy
servicesof Dr HerbertWone at IRL.
References
Ciancia,M., Matulewicz,M. C., Finch, p., & Cerezo,A. S. 0993a). Deter_
mination of the structures of cystocarpic carrageenans from Gigartina
skottsbergii by methylation analysis and NMR spectroscopy. Carbohydrate Research, 238. 241 -249.
Ciancia, M., Noseda, M. D., Matulewicz, M. C., & Cerezo. A. S.
( 1993b).Alkali-modification of carrageenans:
mechanismanclkinetics
in kappa,/iota,mu./nu,and lambda-series. Carbohy,tlrate polymers, 20,
95-98.
Craigie,J. S. (1990).Cell walls. In K. M. Cote & R. C. Sheath,Biology of
the red algae (pp.22l-257). CambridgeUniversity press.
Dolan, T. C. S., & Rees, D. A. (1965). The carrageenansparr II: rhe
positions of the glycosidic linkages and sulfate es(ers in lambdacarrageenan.Joumal of the Chemical Society,3534-3539.
Falshaw,R., & Furneaux,R. H. (1994). Carrageenanfrom the tetrasporic
stageof Gigarrina decipiens.CarbohydrateResearch,252, l7l-182.
Falshaw,R., & Fumeaux, R. H. (1995). The structural analysisof disaccharides from red algal galactâns by methylation and reductive partialhydrolysis. Carbohydrate Resea rc h, 269, 183- l 89.
Falshaw,R., Furneaux,R. H., Wong, H., Bacic, A., Liao, M-L., & Chantlrkrachang,S. (1996). Structuralanalysisof carrageenansfrom Burmese
and Thai samples of Catenella nipae Zanardini. Carbohydrate
Research,285, 8 1-98.
Falshaw,R., Fumeaux,R. H., & Stevenson,D. E. (1998). Agars from nine
species of red seaweedin the genus Curdiea (Gracilariaceae, Rhodophyta). Carbohydrate Research, -108, 10?-I I5.
Jol, C. N., Neiss, T. G., Penninkhof,8., Rudotph, 8., & DeRuiter, G. A.
(1999). A novel high-performanceanion-exchangechromatographic
method for the analysis of carrageenansand agars containing 3,6anhydrogalactose.Analytical Biochemistry, 268, 213-222.
KnutsenS
, . H . , M y s l a b o d s k iD
, . E . , L a r s e n , 8 . , & U s o v , A . f . ( 1 9 9 4 ) .A
modified system of nomenclaturefor red algal galactans.Botanica
Marina. 37. 163-169.
Lawson, C. J., Rees, D. A., Stancioff, D. J., & Stanley, N. F. (1973).
Carrageenans-Part VIil: repeating structures of galactan sulfates
from Furc ellaria fastigiata, G i gartina conaliculata, Gigartina cfutmissoi, Gigartina atropurpurea, Ahnfeltia durvillaei, Gymnogongrus
furcellatus, Eucheuma isifurme, Eucheuma uncinatum, Aghardiella
tenera, Pachymenia hymantophora, and Gloiopelris t:enicomis.
Joumal of the ChemicalSocietyPerkin, I,2l'7'7-2182.
Matulewicz, M. C., Ciancia,M., Noseda,M. D., & Cerezo, A. S. (1989).
Carrageenansystems from tetrasporic and cystocarpic stagesof Gryartina skottsbergii. Phytochemistry, 28 ( I I ), 2937 -2941 .
Matulewicz, M. C., Ciancia, M., Noseda, M. D., & Cerezo, A. S.
(1990).Methylationanalysisof canageenansfrom tetrasporophytic
and
cystocarpic stages of Gigarrina skonsbergii. Phytr.tchemisrry,29 (ll).
3 4 0 7 , 3 4 10 .
in
McCandless,E. L., Craigie,J. S., & Walter, J. A. (1973). Carrageenans
the gametophyteand sporophytic stagesof Cfutndrus crispus. Plunla
(Berlin), I I 2, 201-212.
Miller, I. J., Falshaw, R., & Furneaux, R. H. (1996). A polysaccharide
fraction from the red seaweed Chantpia nowe-iealandiae, Rhodomeniales,Rhodophyta.Hydrobiologia, 326/327, 505-509.
N u n a n ,K . J . , S i m s , I . M . , B a c i c , A . , R o b i n s o n ,S . P . , & F i n c h e r ,G . B .
( 1 9 9 8 ) .C h a n g e si n c e l l w a l l c o m p o s i t i o nd u r i n g r i p e n i n g o f g r a p c
beries. Plant Phtsiologv, I I 8,'783-192.
PenmanA
, . . & R e e s ,D . A . ( 1 9 7 3 ) .C a r r a g e e n a n s - P a r tI X : r n e t h y l a t i o n
Oigurtinu
analysis of galrctan sulfates liorn F'urcelluria .lQSti!:iQrd,
+tl
R. Fulshaw et a!. / Food Hydror:olloùls (2001) ,t4l
*152
l5
( unal i cu l o ta, () i ga rt i na chanùs.çoi,
C i go r t ina al ropu rpu rea, A trnfel t ia
durv'illaei, cyntnogongrus
furceilatus, Encheunn'isif)me, Euchcuma
unr.inaturrt. Aghurdiellu tenera, pachymenia
hynnntophora, anJ
Gloiopeltis cen,icrsmis. Structure of
{-carrageenan- Joumal of the
ChenticulSocieh-,perkin, I, Zlg2_Zlg7.
Pickrnere.S. E., Parsons.M. J., & Bailey, R. W. (1973).
Composition of
Gigartina carrageenanin relation to sporophyte and gametophyte
stages
of the life cycle. Phr-tochemistry,12,2441_2444.
Rees,_D.A. (1972). Shapely polysaccharides.Biochemistry
.126,
Joumal,
)\7 _)'71,
Stevenson,T. T., & Furneaux,R. H. (1991). Chemical merhods
for the
analysis of sulphatedgalactansfrom red algae. Carbohydrate
Research,
-298.
2 t0, 2'17
S t o r t z ,C . A . , & C e r e z o ,A . S . ( 1 9 9 2 ) . T h c r r c N M R
s p e c t r o s c o poy l
carrageenans:
calculationof chemical shifts and cornrrutcr_aided
struc_
tural detennination.Carbohy,lrutcpolttncrs, I 5,217 _242.
Stortz, C. A., & Cerezo,A. S. (1993). The systemsofcanageenanslrom
cystocarptc and tetrasporic stagesfrom Iridaca undulosu. Carbohvdrtrtt
Research,242. 217-2Zi .
Stonz,C. A., Bacon,B. E., Chemiak, R., & Cerezo,A. S. ( 1994).High_field
NMR spectroscopy of cystocarpic and tetrasporic canageenans from
Iridaea undulosa. Carbohydrate Research, 261 , 3ll. _326.
Usov, A. I., & Shashkov,A. S. (1985). polysaccharidesof algaeXXXIV:
detection of iota-carra-geenanin phyllophora brodittei (Tum.) J.Ag.
(Rhodophyta) using r3C-NMR spectroscopy. Botanica Marina, 2g,
367-373.