-
1746 JOURNAL OF FOOD SCIENCEVol. 67, Nr. 5, 2002 2002 Institute
of Food Technologists
Food Chemistry and Toxicology
JFS: Food Chemistry and Toxicology
Hydrolysis of Caseins By Extracts of CynaraCardunculus
Precipitated by Ammonium SulfateS.V. SILVA, R.M. BARROS, AND F.X.
MALCATA
ABSTRACT: Polyacrylamide gel electrophoresis (in the presence of
urea) and gel permeation chromatographywere employed to assess the
profile of hydrolysis of caseins and the activity of enzymes
contributed by the flowersof the plant Cynara cardunculus on bovine
caseins, after previous precipitation with ammonium sulfate or in
aplain crude aqueous extract. Results indicated that the
qualitative degradation profile of bovine caseins by plantenzymes
(cardosins) remains essentially unchanged upon extraction, and
quantitative analysis showed that theprecipitated fractions had a
higher coagulant-to-proteolytic activity ratio; hence, the results
showed that inex-pensive precipitation with ammonium sulfate can
successfully be used as a purification method in the productionof
that plant coagulant in standardized form.
Keywords: milk proteins, thistle, FPLC, urea-PAGE
Introduction
CRUDE AQUEOUS EXTRACTS OF FLOWERS OF THE COMMON WILDthistle,
Cynara cardunculus, have been used for ages as a co-agulant in the
manufacture of several traditional Portugueseand Spanish cheeses,
such as Serra da Estrela, Serpa, andAzeito (Vieira de S and Barbosa
1972; Roseiro 1991; Macedoand Malcata 1997), and Los Pedroches, La
Serena, and Flor deGua (Fernandez-Salguero and others 1991; Sanjuan
andFernandez-Salguero 1994). Those flowers are known to contain
2aspartic proteinases, currently termed cardosins A and B,
eachconsisting of 2 subunits with apparent molecular weights of
31and 15 kDa, and 34 and 14 kDa, respectively; it has been
claimed(Pires and others 1994; Verssimo and others 1995) that
cardosinA is similar to chymosin, while cardosin B is similar to
pepsin, interms of specificity and activity.
Those enzymes are currently purified by a 2-step procedure,which
involves liquid extraction at low pH and separation after-wards by
liquid chromatography (gel filtration followed by ion-exchange)
(Verssimo and others 1995). In addition to being cost-ly, the
overall process is rather slow and cumbersome; hence, it ishardly
appropriate when large quantities of those enzymes aresought. The
objective of enzyme purification is to remove non-protein
contaminants, as well as to isolate the enzyme in ques-tion from
other proteins; the former objective is relatively easy toachieve,
unlike the latter. Precipitation techniques can be usedat an early
point, whereas chromatographic procedures (for ex-ample,
ion-exchange, gel filtrarion, or adsorption chromatogra-phy) are
usually employed later. Salting-out is the most commonprecipitation
technique (Picn and others 1994). The salt usuallychosen is
ammonium sulfate, (NH4)2SO4, which is highly solublein water (Robyt
and White 1990) and is known to effectively sta-bilize proteins
(Scopes 1994). Since large amounts of salt will con-taminate the
precipitated proteins, removal thereof prior to en-zyme usage is
required; this can easily be achieved by dialysis orgel filtration
(Scopes 1994). In an early approach to this issue,Sousa and Malcata
(1996) have studied the effects of processingconditions on the
caseinolytic activity of crude extracts of C. car-dunculus, and
found that the maximum specific caseinolytic ac-tivity was obtained
by grinding the flowers for 36 s, using an ex-
traction buffer of pH 5.9 and NaCl content of 0 % (w/w), and
ho-mogenizing the ground flower/buffer suspension for 15 min.
The objective of this study was (i) to evaluate the
proteolyticand coagulant activities of said plant enzymes, both in
crudeaqueous extract and following precipitation with ammonium
sul-fate at 2 concentration levels (quantitative action), and (ii)
tocharacterize the effect of those enzymes on the profile of
hydrol-ysis of bovine s- and -caseins (qualitative action). The
workinghypothesis under scrutiny was to ascertain whether
salting-outmight replace the (more expensive) classically employed
chro-matographic separations.
Material and Methods
Partial purification of enzymesThe source of plant proteases was
dry flowers of C. carduncu-
lus: the stigmata and stylets of dry flowers were macerated
with0.1 M aqueous citric acid (pH 3.0) at the ratio of 1 gflowers
per 10mLbuffer, so as to produce a fluid product. Ammonium sulfate
wasthen added to the plant extract up to 30% (w/v) saturation.
After30 min, the solution was centrifuged at 10000 rpm for 10 min
at4 C; the precipitate was redissolved in water up to
approximate-ly twice the pellet volume. Ammonium sulfate was added
onceagain to the supernatant up to 70% (w/v) saturation. After
30min, the solution was centrifuged at 10000 rpm at 4 C; as
before,the precipitate was redissolved in twice its volume of
water. Plantenzyme extracts were dialyzed overnight at 4 C against
plain wa-ter, and lyophilized prior to use.
Electrophoretic characterization of enzymesPredetermined amounts
of each cardosin extract were placed
in separate Eppendorf vials, and 100 mL of 10% (w/w) SDS
wasadded to each one. The vials were heated at 90 C for 10 min in
aheating block, and then cooled to room temperature. The
sepa-ration was carried out in high-density gels: each gel had a
stack-ing gel zone (7.5% T, 2% C) and a continuous separating gel
zone(20% T and 2% C); they also contained 30% (v/w) ethylene
glycol.The automated electrophoresis system Phastsystem
(Pharma-cia, Uppsala, Sweden) was used to separate the proteins,
which
jfsv67n5p1746-1751ms20010529-SR.p65 7/8/2002, 2:21 PM1746
-
Vol. 67, Nr. 5, 2002JOURNAL OF FOOD SCIENCE 1747
Food
Chem
istry
and T
oxico
logy
Hydrolysis of caseins by plant enzymes
were stained by Coomassie blue.
Coagulant activityThe clotting activity was determined following
the procedure
described by IDF (1992). The reconstituted milk was obtained
bydissolving low-heat skim milk (NILACTM) (NIZO, Ede, The
Neth-erlands) in 10 mM aqueous CaCl2 (pH 6.5), so as to achieve a
con-centration of 0.12 kg/L; the time taken for 0.2 mL of each
enzymeextract to coagulate 2 mL of reconstituted milk was
measuredand recorded (the coagulation point was determined by
manual-ly rotating the test tube periodically, at short time
intervals, andchecking for visible clot formation). One rennet unit
(R.U.) wasdefined as the amount of protein that coagulates 10 mL of
recon-stituted low-heat skim milk powder at 30 C in 100 s
(Berridge1945).
Proteolytic activityThe proteolytic activity was determined by
the method of
Tomarelli and others (1949), with slight modifications. The
lyo-philized enzyme mixture, reconstituted in 0.1 M aqueous
sodiumphosphate buffer (pH 6.0), was mixed with 250 L of 2%
(w/v)azocasein and incubated at 25 C for 10 min. The reaction
wasthen quenched via addition of 0.5 mL of cold 5% (w/v)
trichloro-acetic acid, TCA. The solution was centrifuged at 10000
rpm for10 min; 1 mL of supernatant was further mixed with 1 mL of
0.5 Maqueous NaOH to intensify the azo-associated color. The
absor-bance of the solution at 440 nm was measured against a
blank,prepared in a similar fashion but adding TCA to azocasein
priorto addition of the enzyme.
Hydrolysis of substratesCommercial bovine s- and -caseins
(Pharmacia) were dis-
solved up to 1 mg/mL in aqueous phosphate buffer (pH
6.5),containing sodium azide (0.05 g/mL) to prevent microbial
activi-ty; these solutions were treated independently with
cardosinsprecipitated by ammonium sulfate at 30% (w/v) and 70%
(w/v)saturation, as well as with the crude enzyme extract, and
incu-bated at 30 C; the (enzyme-containing) protein:substrate
ratioused in all cases was 1:400 (w/w). At predetermined time
inter-vals (0 min, 30 min, 1 h, 2 h, 5 h, 8 h, 12 h, and 24 h),
sampleswere taken and the reaction was quenched via addition of
dou-ble-concentrated buffer at 50%(v/v) (McSweeney and others1993)
in the case of samples for electrophoresis (urea-PAGE), orvia
heating at 95 C for 15 min in the case of samples for fast pro-tein
liquid chromatography (FPLC).
Electrophoretic characterization of hydrolysatesSamples (0.75
mL) of hydrolysates were obtained, and pre-
pared for urea-PAGE as described above. Urea-PAGE (12.5% forthe
separation gel and 4% for the stacking gel; pH 8.9) was per-formed
following the procedure of Andrews (1983), with themodifications
proposed by Shalabi and Fox (1987). Gels werestained with Coomassie
Blue G250 (Bio Rad Laboratories, Her-cules, Calif., U.S.A.)
according to the method of Blakesley andBoezi (1977).
FPLC characterization of hydrolyzatesFPLC was conducted
according to the following protocol: sep-
aration was performed after injection of 100 L of each
hydroly-sed sample in a Superose 12 column HR 10/30 (Pharmacia),
withthe aid of an injection valve MV-7 and a UV-MII-detector
(operat-ed at 280 nm); the mobile phase was 150 mM of aqueous NaCl
in0.05 M aqueous phosphate buffer (pH 7.0), containing 0.2 g/L
NaN3 as preservative, for 80 min at a flow rate of 0.4 mL/min.
Pri-or to injection, the sample was passed through a 0.45 m
filter(Nucleopore, Cambridge, Mass., U.S.A.), whereas the buffer
wasfiltered through 0.22 m paper filter (Nalgene, Rochester,
N.Y.,U.S.A.). The void volume of the column was determined
previ-ously using blue dextran. The retention times of the peaks
ob-tained were compared with those of a mixture of molecularweight
standards, namely aldolase (158 kDa), bovine serum al-bumin (67
kDa), ovalbumin (43 kDa), -lactoglobulin (36 kDa),-lactalbumin
(14.4 kDa), and ribonuclease (13.7 kDa). Each an-alytical
determination was carried out in duplicate.
Results and Discussion
Characterization of enzymesThe salting-out technique was
employed to purify the aspar-
tic proteinases present in the crude aqueous extract of C.
cardun-culus; a 1st fraction was precipitated at 30% (w/v)
saturation withammonium sulfate and a 2nd one at 70% (w/v)
saturation withthe same salt. The protein concentration was 92,
197, and 410mgprotein/ lyophilizate for the crude extract, the 70%
(w/v) precipi-tate and the 30% (w/v) precipitate, respectively.
SDS-PAGE wasperformed on the purified enzyme fractions, as well as
on thecrude aqueous extract (Figure 1). The electrophoretogram
re-vealed that precipitates were qualitatively similar to each
other,and that the material in all fractions migrated as 4 bands,
withmolecular weights 31 and 15 kDa (accounted for by cardosin
A),and 34 and 14 kDa (accounted for by cardosin B) (Verssimo
andothers 1995). It is worth noting that 2 extra bands, with
molecularweights 20 to 26 kDa, were present in the fraction
correspondingto the enzymes purified at the lower level of addition
of ammoni-um sulfate. These 2 bands may likely be accounted for by
con-taminating proteins that co-precipitate at 30% (w/v)
saturationof ammonium sulfate, although the possibility that
multimericnative proteins dissociate into their monomeric subunits
duringthe 1st stage of purification can not be completely ruled
out.
Table 1 shows the specific enzymatic activity of the
precipi-
Figure 1SDS-PAGE electrophoretogram of fractions pre-cipitated
by ammonium sulfate. Lane 1molecular weightmarkers, lane 2crude
aqueous extract, lane 3super-natant at 70% (w/v) saturation, lane
4precipitate at 30%(w/v) saturation, lane 5precipitate at 70% (w/v)
satura-tion.
jfsv67n5p1746-1751ms20010529-SR.p65 7/8/2002, 2:21 PM1747
-
Food Chemistry and Toxicology
1748 JOURNAL OF FOOD SCIENCEVol. 67, Nr. 5, 2002
Hydrolysis of caseins by plant enzymes
tates as compared with that of the crude extract. In what
pertainsto the ammonium sulfate-precipitated fractions, the
precipita-tion at 70% (w/v) saturation led to a decrease in the
specific coag-
ulant activity relative to that at 30% (w/v), whereas the
oppositeheld for the proteolytic activity. In general,
fractionation caused
Figure 2Urea-PAGE electrophoretograms illustrating
thedegradation patterns of bovine -casein (-CN) by
cardosinspreviously precipitated with ammonium sulfate at (a)
30%(w/v) and (b) 70% (w/v) saturation and (c) by crude
extract,after incubation for 0, 0.5, 1, 2, 5, 8, 12 and 24 h (lanes
1-8, respectively). Lanes 9 and 10 represent -casein
afterincubation for 0 and 24 h, respectively, in the absence
ofenzyme.
Figure 3Urea-PAGE electrophoretograms illustrating
thedegradation patterns of bovine -casein (-CN) by
cardosinspreviously precipitated with ammonium sulfate at (a) 30%
(w/v) and (b) 70% (w/v) saturation and (c) by crude extract,after
incubation for 0, 0.5, 1, 2, 5, 8, 12 and 24 h (lanes
18,respectively). Lanes 9 and 10 represent s-casein after
incu-bation for 0 and 24 h, respectively, in the absence of
enzyme.
jfsv67n5p1746-1751ms20010529-SR.p65 7/8/2002, 2:21 PM1748
-
Vol. 67, Nr. 5, 2002JOURNAL OF FOOD SCIENCE 1749
Food
Chem
istry
and T
oxico
logy
Hydrolysis of caseins by plant enzymes
a substantial decrease of the specific proteolytic activity (P)
anda less pronounced decrease in the specific clotting activity
(C),both relative to that of the crude extract. A decrease of the
totalproteolytic activity of purified fractions of sodom apple
(Calotro-pis procera) relative to the initial extract was also
reported byAworh and Nakai (1986). Chen and Zall (1986a,b) claimed,
inturn, an increase of the clotting activity in a fraction of clam
vis-cera extracts fractionated with ethanol. The ratio of
coagulant-to-proteolytic activity is a useful indicator of the
enzyme appropri-ateness for general use in cheesemaking: a higher
C/P ratiousually leads to higher quality cheeses (Chen and Zall
1986b;Harboe 1991; Esteves 1995). The ratio was actually higher for
thefractions precipitated with ammonium sulfate, as compared
withthe plain, crude aqueous extract; between the precipitated
frac-tions, the ratio was much higher for the 30% (w/v) saturation
pre-cipitate.
Electrophoretic characterization of hydrolysatesThe progress of
hydrolysis of bovine -and s-caseins brought
about by cardosins precipitated at the 2 levels of ammonium
sul-fate and by crude aqueous extract is shown in Figure 2 and 3,
re-spectively.
Quantitative breakdown of original -casein was observed af-ter
as early as 30 min, irrespective of the level of ammonium sul-fate
added for precipitation of enzyme, whereas at least 5 h hadto
elapse before a similar result was achieved with the crude
ex-tract. In the hydrolysates, -I-casein was the 1st peptide to
showup in all 3 situations tested, which is presumably a result
ofcleavage of Leu192-Tyr193 in bovine -casein; according toSimes
(1998), that is the peptide bond in pure bovine -caseinmost
susceptible to the action of either cardosin. The peptide
(-I-casein) was further hydrolyzed to -II- and -III-casein,
ac-cording to the classification proposed by Creamer (1976).
Notethat Visser and Slangen (1977) observed that hydrolysis of
pure-casein by chymosin results in the production of 3
peptides,namely -I-casein (f1-189/192), -II-casein
(f1-163/165/167),and -III-casein (f1-127/139).
The hydrolysis of s-casein was essentially complete by 30min,
regardless of the level of ammonium sulfate added for
theprecipitation of enzyme, whereas at least 2 h had to elapse
be-fore an identical situation was produced by the crude
extract.One of the fast moving peptide bands was assigned to
s1-I-casein, following comparison with the electrophoretic mobility
ofbovine s1-I-casein produced via action of chymosin (Mulvihilland
Fox 1977); such peptide appeared in all 3 situations after 1min of
incubation. Those peptides were subsequently degradedto smaller
peptides, which migrated close to each other underthe electric
field. Ramalho-Santos and others (1996) have stud-
Table 1Specific coagulant and proteolytic activities,
andcorresponding ratio, of ammonium sulfate-precipitatedfractions
and crude extract.
Coagulant Proteolyticactivity (C) activity (P)
Fraction (R.U./gprotein) (DAbs/gprotein/min) C/P ratioCrude
extract 104 4.11 1.53 0.10 67.7 10.3030% (w/v) 90.6 2.01 0.24 0.001
377.5 34.62 saturation70% (w/v) 74.6 2.69 0.79 0.004 94.4 11.64
saturationNote: Abs AbsorbanceValues represent means of 3
replicates confidence interval (p = 95%)Data referred to dry
basis
ied the independent action of cardosins A and B on
bovines1-casein, and reported that both enzymes have preference
forPhe23-Phe24, but are able to cleave bonds Trp164-Tyr165
andPhe153-Tyr154, although to lower extents; Phe23-Phe24 was
alsofound by Mulvihill and Fox (1977) to be one of the peptide
bondsmost labile to the action of chymosin in the pH range
2.2-7.0.
Consequently, both caseins were fully degraded by the
car-dosins, and followed similar degradation patterns in either
crudeor partially separated form. However, the rate of hydrolysis
by
Figure 4Molecular weight distribution of peptides pro-duced via
hydrolysis of bovine -casein by cardosins pre-viously precipitated
with (a) 30% (w/v) and (b) 70% (w/v)saturation of ammonium sulfate
and by (c) crude extract,after incubation for 0, 0.5, 1, 2, 5, 8,
12 and 24 h.
jfsv67n5p1746-1751ms20010529-SR.p65 7/8/2002, 2:22 PM1749
-
Food Chemistry and Toxicology
1750 JOURNAL OF FOOD SCIENCEVol. 67, Nr. 5, 2002
Hydrolysis of caseins by plant enzymes
the crude extract was lower than by the partially purified
ex-tracts, as expected. Note that the
(enzyme-containing)protein:substrate ratio used was the same in the
reactions of hy-drolysis of both substrates, but that the amount of
enzymepresent was higher in the ammonium sulfate-saturated
fractionthan in the crude extract (see Figure 1); this may help
explain ourresults, because the lower enzyme content, the lower
rate of hy-drolysis. The rate of degradation of s-casein was also
slightlyhigher than that of -casein when the crude extract was
consid-ered. This finding is in agreement with the reported fact
(Mulvi-
hill and Fox 1977) that -casein is less susceptible than
s1-caseinto proteolysis by chymosin. Sousa and Malcata (1997) also
re-ported that -casein undergoes less degradation than s1-caseinin
bovine milk cheeses manufactured with aqueous extracts offlowers of
C. cardunculus.
Chromatographic characterization of hydrolysatesThe molecular
weight (MW ) distribution of peptides in the
casein hydrolysates is shown in Figures 4 and 5 for - and
s-caseins, respectively. For both caseins, hydrolysis was in all
casesaccompanied by a decrease in the concentration of
high-MWpeptides (20 to 25 kDa) and a concomitant increase in the
con-centration of low-MW ones (< 1 kDa and 1 to 2.5 kDa). This
obser-vation reflects the expected sequential hydrolysis of longer
toshorter peptides. No relevant changes in the concentration
ofpeptide material between 2.5 and 15 kDa was noticed over the 24h
of hydrolysis in all cases, which indicates somewhat unselec-tive
hydrolysis.
As far as -casein is concerned, the family of peptides withMW
between 20 and 25 kDa essentially vanished by 2 h of hy-drolysis
carried out by the purified extracts (Figure 4a and b).The 1st
short peptides to be noticed were those with MW in therange of 1 to
2.5 kDa, after as little as 30 min of reaction; then thepeptide
group with MW within 15 to 20 kDa started to build up.The high-MW
peptide group was present until 12 h of hydrolysisby the crude
extract, but gradually yielded peptides with MW of15 to 20 kDa
(Figure 4c).
Regarding s-casein, peptides with MW of 20 to 25 kDa stillshowed
up by 2 h of hydrolysis by the crude extract (Figure 5c);however,
that group disappeared by 30 min of reaction when thepartially
purified extracts were used (Figure 5a and b). In all 3
sit-uations, peptides with MW of 15 to 20 kDa and below 2.5 kDa
ap-peared simultaneously with disappearance of the peptide
groupwith MW in the range 20 to 25 kDa.
Conclusions
THE DEGRADATION PATTERNS OF S- AND -CASEIN ARE SIMILAR,
IR-respective of the extent of purification of the plant
extracttested. However, when crude extract is used, the rate of
hydroly-sis is lower than when precipitation by ammonium sulfate
tookplace previously; the crude extract possesses indeed a lower
ratioof coagulant (C) to proteolytic (P) activities. Salting-out of
car-dosins is apparently appropriate for the production of more
effi-cient coagulants for cheese manufacture in terms of higher
C/Pratio, and is thus preferable to classical (more expensive)
chro-matographic steps of purification.
ReferencesAndrews AT. 1983. Proteinases in normal bovine milk
and their actions on
caseins. J Dairy Res 50(1):45-55.Aworh OC, Nakai S. 1986.
Extraction of milk clotting enzyme from sodom apple
(Calotropis procera). J Food Sci 51(6):1569-1570.Berridge, NJ.
1945. Some observations on the determination of activity of
ren-
net. Analyst 77:57-72.Blakesley RW, Boezi JA. 1977. A new
staining technique for proteolysis in poly-
acrylamide gels using Coomassie Brilliant Blue G250. Anal
Biochem 82:580-582.
Chen HC, Zall RR. 1986a. Partial purification and
characterization of cathepsinD-like and B-like acid proteases from
surf clam viscera. J Food Sci 51(1):71-75.
Chen HC, Zall RR. 1986b. Evaluation of thiol-activated proteases
from clam vis-cera as rennet substitute for cheese-making. J Food
Sci 51(3):815-820.
Creamer LK. 1976. A further study of the action of rennin on
-casein. NZ J DairySci Technol 11(1):30-39.
Esteves CLC. 1995 Estudo comparativo das caractersticas
bioqumicas dos coal-hos de Cynara cardunculus L., Cynara scolymus
L. e Cynara humilis L. [MScthesis]. Coimbra, Portugal: Univ. de
Coimbra. 84 p.
Fernandez-Salguero J, Sanjun E, Montero E. 1991. A preliminary
study of thechemical composition of Gua cheese. J Food Comp Anal
4(3):262-269.
Harboe MK. 1991. Chymogen, a chymosin rennet manufacture by
fermentationof Aspergillus niger. Bull Int Dairy Fed 269:3-7.
Figure 5Molecular weight distribution of peptides pro-duced via
hydrolysis of bovine -casein by cardosins pre-viously precipitated
with (a) 30% (w/v) and (b) 70% (w/v)saturation of ammonium sulfate
and by (c) crude extract,after incubation for 0, 0.5, 1, 2, 5, 8,
12 and 24 h.
jfsv67n5p1746-1751ms20010529-SR.p65 7/8/2002, 2:22 PM1750
-
Vol. 67, Nr. 5, 2002JOURNAL OF FOOD SCIENCE 1751
Food
Chem
istry
and T
oxico
logy
Hydrolysis of caseins by plant enzymes
IDF. 1992. Bovine rennets. Determination of total milk-clotting
activity. IDFStandard 157. Brussels, Belgium: International Dairy
Federation.
Macedo AC, Malcata, FX. 1997. Hydrolysis of s-and -caseins
during ripeningof Serra cheese. Food Chem 58(1-2):43-48.
McSweeney PLH, Olson NF, Fox PF, Healy A, Hojrup P. 1993.
Proteolytic specificityof chymosin on bovine s1-casein. J Dairy Res
60(3):401-412.
Mulvihill DM, Fox PF. 1977. Proteolysis of s1-casein by
chymosin: influence of pHand urea. J Dairy Res 44(3):533-540.
Picn A, Gaya P, Medina M, Nuez M. 1994. Microencapsulation of
cyprosinsfrom flowers of Cynara cardunculus L. in
dehydration-rehydration liposomes.Biotechnol Lett
16(10):1031-1034.
Pires E, Faro C, Macedo I, Esteves C, Morgado J, Verssimo P,
Dias-Pereira, GomesD. 1994. Flor do cardo versus quimosina no
fabrico de queijos artesanais. RevSoc Port Qum 54:66-68.
Ramalho-Santos M, Verssimo P, Faro C, Pires E. 1996. Action on
bovine s1-caseinof cardosins A and B, aspartic proteinases from the
flowers of the cardoonCynara cardunculus L. Biochim Biophys Acta
1297(1):83-89.
Robyt JF, White BJ. 1990. Biochemical TechniquesTheory and
Practice. ProspectHeights: Waveland Press. 408 p.
Roseiro MLB. 1991. Ewes milk cheesemaking in Portugal using a
vegetable ren-net (a review). Sheep and Dairy News 8:65-80.
Sanjuan E, Fernandez-Salguero J. 1994. Influence of some factors
on the coag-ulation time by vegetable coagulant (Cynara sp.).
Alimentacion, Equipos yTecnologa 13(6):69-73.
Scopes RK. 1994. Protein PurificationPrinciples and Practice.
New York: Spring-er-Verlag. P 380.
Shalabi SI, Fox PF. 1987. Electrophoretic analysis of cheese:
comparison of meth-ods. Irish J Food Sci Technol 11(2):135-151.
Simes IIG. 1998. Caracterizao molecular da aco das cardosinas A
e B sobrecasenas e bovinas [MSc thesis]. Coimbra, Portugal: Univ.
de Coimbra. 112p.
Sousa MJ, Malcata FX. 1996. Effects of processing conditions on
the caseinolyticactivity of crude extracts of Cynara cardunculus L.
Food Sci Technol Int 2:255-263.
Sousa MJ, Malcata FX. 1997. Comparative biochemical evolution
during ripen-ing of bovine, ovine and caprine cheeses manufactured
with exracts of flowersof Cynara cardunculus. Z Lebensm Forsch A
205:97-103.
Tomarelli R, Charney M, Harding M. 1949. The use of azoalbumin
as a substratein the colorimetric determination of peptic and
tryptic activity. J Lab Clin Med34:428-433.
Verssimo P, Esteves C, Faro C, Pires E. 1995. The vegetable
rennet of Cynaracardunculus L. contains 2 proteinases with chymosin
and pepsin-like speci-ficities. Biotechnol Lett 17(6):621-626.
Vieira de S F, Barbosa M. 1972. Cheese-making with a vegetable
rennet fromCardo (Cynara cardunculus). J Dairy Res
39(3):335-343.
Visser S, Slangen J. 1977. On the specificity of chymosin
(rennin) in its action onbovine -casein. Neth Milk Dairy J
31(1):16-30.
MS 20010529 Submitted 9/25/01, Accepted 12/4/01, Received
12/4/01
Financial support for authors Silva and Barros was provided by
PhD fellowships (BD/18479/98 and BD/16037/98, respectively), issued
by program PRAXIS XXI (FCT, Portugal).
Authors are with the Escola Superior de Biotecnologia, Univ.
CatlicaPortuguesa, Rua Dr. Antnio Bernardino de Almeida, P-4200-072
Porto, Por-tugal. Direct inquiries to author Malcata (E-mail:
[email protected]).
jfsv67n5p1746-1751ms20010529-SR.p65 7/8/2002, 2:22 PM1751