Plant Physiology and Biochemistry 52 (2012) 104e111

Contents lists available at SciVerse ScienceDirect

Plant Physiology and Biochemistry

journal homepage: www.elsevier.com/locate/plaphy

Research article

Purication and physicochemical characterization of a serine protease with

brinolytic activity from latex of a medicinal herb Euphorbia hirta
Girijesh Kumar Patel, Ashish Ashok Kawale, Ashwani Kumar Sharma*
Department of Biotechnology, Indian Institute of Technology Roorkee, Roorkee 247 667, India

a r t i c l e i n f o

a b s t r a c t

Article history:
Received 9 November 2011
Accepted 7 December 2011
Available online 13 December 2011

A 34 kDa serine protease, designated as hirtin, with brinolytic activity was puried to homogeneity
from the latex of Euphorbia hirta by the combination of ion exchange and gel ltration chromatography.
The N-terminal sequence of hirtin was found to be YAVYIGLILETAA/NNE. Hirtin exhibited esterase and
amidase activities along with azocaseinolytic, gelatinolytic, brinogenolytic and brinolytic activities. It
preferentially hydrolyzed Aa and a-chains, followed by Bb and b, and g and geg chains of brinogen and
brin clot respectively. The optimum pH and temperature for enzyme activity was found to be pH 7.2 and
50
C respectively. Enzymatic activity of hirtin was signicantly inhibited by PMSF and AEBSF. It showed
higher specicity for synthetic substrate p-tos-GPRNA for thrombin. The CD spectra of hirtin showed
a high content of b-sheets as compared to a-helix. The results indicate that hirtin is a thrombin-like
serine protease and may have potential industrial and therapeutic applications.
2011 Elsevier Masson SAS. All rights reserved.

Keywords:
CD studies
Euphorbia hirta
Euphorbiaceae
Fibrinogenolytic activity
Fibrinolytic activity
Kinetic studies

1. Introduction
Blood coagulation and brinolysis, involving a series of serine
proteases, are two important processes associated with haemostasis and wound healing [1]. Fibrinogen is a heterotrimer protein
having three subunits namely Aa, Bb and g. The Aa and Bb subunits
are specically hydrolysed by the thrombin resulting the insoluble
brin clot formation (a, b and geg) at the nal step of blood
coagulation in response to an injury [2]. Fibrinolysis dissolves the
brin clot through the action of plasmin resulting in wound healing. Fibrin clot formation and brinolysis are tightly regulated
processes. However, thrombotic disorders may occur in a condition
when brin hydrolysis does not take place. Intravascular thrombosis resulting from brin aggregation in arteries is the main cause
of cardiovascular disorders including myocardial infarction [3]. The

Abbreviations: BSA, bovine serum albumin; BAEE, N-benzoyl-L-arginine ethyl

ester; BTEE, N-benzoyl-L-tyrosine ethyl ester; BAPNA, N-benzoyl-L-arginine-pnitroanilide; BTNA, N-benzoyl-L-tyrosine p-nitroanilide; p-tos-GPRNA, N-p-tosyl-LGly-L-Pro-L-Arg p-nitroanilide; AAPF, N-succinyl-L-Ala-L-Ala-L-Pro-L-Phe-p-nitroanilide; AAPL, N-succinyl-L-Ala-L-Ala-L-Pro-L-Leu-p-nitroanilide; AAA, N-succinyl-LAla-L-Ala-L-Ala-p-nitroanilide; AAV, N-succinyl-L-Ala-L-Ala-L-Val-p-nitroanilide;
PMSF, phenylmethylsulfonyluoride; AEBSF, Aminoethyl-benzene sulfonyl uoride
hydrochloride; EDTA, Ethylenediaminetetraacetic acid; IAA, Iodoacetic acid; E64, 1trans-epoxysuccinylleucylamide(4-guanidino)butane-N-[N-(L-3-trans-carboxyirane-2-carbonyl)-L-leucyl] agimatine; SDS, Sodium dodecyl sulphate; CBB R-250,
Coomassie brilliant blue R-250.
* Corresponding author. Tel.: 91 1332 285657; fax: 91 1332 273560.
E-mail addresses: aksbsfbs@yahoo.co.in, aksbsfbs@iitr.ernet.in (A.K. Sharma).
0981-9428/$ e see front matter 2011 Elsevier Masson SAS. All rights reserved.
doi:10.1016/j.plaphy.2011.12.004

brinolytic agents currently available for treatment of thrombosis

are plasminogen activators such as tissue-type plasminogen activators and urokinase-type plasminogen activators and all these
agents exhibit undesirable side effects. Therefore, the search for
other brinolytic enzymes as therapeutic agents for treatment of
thrombotic disorders from diverse sources is needed. Previously,
many proteases affecting coagulation and brinolysis have been
studied from various sources including plants [4e6], annelids [7,8],
insects [9], snake venom [10,11], bacteria [12] and recently from
Korean mushroom [13].
Plant latex is rich source of multiple proteases belonging mainly
to cysteine or serine super family [5,14]. They have been reported
from different plant families including Asclepiadaceae, Apocynaceae,
Moraceae, and Euphorbiaceae [15]. Earlier, many proteases from the
latex of different species of Euphorbia genus have been reported
[16]. In recent years, proteases affecting coagulation and brinolysis have been isolated and characterized from plant latex particularly from Euphorbiaceae family [4,14]. Traditionally, the lattices
from different plants have been used to stop bleeding which has
been attributed to presence of proteases having ability to affect
blood coagulation.
Euphorbia hirta, belonging to Euphorbiaceae family, is a small,
herbaceous wild plant with great medicinal value. Traditionally, the
plant has been used for the treatment of respiratory ailments and
gastrointestinal disorders [17]. Many organic compounds have
been isolated and characterized from this plant such as avonoids,
polyphenols, tannins, triterpenes, phytosterols [18]. It is reported
that the plant extract have anti-allergic, anti-inammatory [19],

G.K. Patel et al. / Plant Physiology and Biochemistry 52 (2012) 104e111

105

antimicrobial [20], diuretic and antidiabetic activity [21]. Till date,

no protein has been characterized and this is the rst protein reported from this plant. The present work describes the purication
and characterization of a novel thrombin-like serine protease,
designated as hirtin, with brinolytic and brinogenoltic activities
from the latex of E. hirta.

2. Result and discussion

2.1. Purication and SDS-PAGE analysis of hirtin
A new brinolytic serine protease named as hirtin was puried
from the latex of E. hirta by the combination of Q sepharose anion
exchange and Superdex 200 gel ltration chromatography. The
crude milky latex, full of the gum and waxy contents, was frozen
at 20
C for 20 h and thawed on ice and centrifuged. The cleared
supernatant of latex, after centrifugation and ltration with
0.45 mm syringe lter, was used as crude enzyme solution and
loaded on to a pre-equilibrated Q sepharose fast ow column.
Bound proteins were eluted with a linear salt gradient of 0e0.5 M
NaCl. The fractions having proteolytic activity were eluted at
150 mM NaCl in 50 mM potassium phosphate buffer pH 7.2 with
slight impurity. For further purication and molecular weight
determination, dialysed and concentrated protein sample was
loaded onto a pre-equilibrated HiLoad 16/60 Superdex 200 column
(16  600 mm, GE Healthcare). The protein was eluted with the
same buffer at a ow rate of 0.5 mL min1 with the retention
volume of 75.25 mL. The protein was concentrated and stored for
further use. The purication results of hirtin are summarized
(Table 1). The activity yield of the puried enzyme was calculated to
be 38.4% with the specic activity of 8.24 activity units mg1 min1
under the optimal assay conditions using BAEE as a substrate. The
fold purication of enzyme is less (3.39) than the expected value
which may be due to presence of the other proteases in the crude
extract of latex and also due to the loss of the activity during
purication, dialysis and concentration of hirtin. The fold purication and yield of hirtin are comparable to other serine proteases
like Dubiumin, Milin and Cryptolepain [22e24]. The relative
molecular mass and purity of hirtin was analysed on 12% SDS-PAGE
under reducing condition. The apparent molecular mass was found
to be 34 kDa on SDS-PAGE (Fig. 1a) while on gel ltration column,
the molecular weight was determined to be 37.4 kDa by comparing
with gel ltration molecular weight standards. SDS-PAGE and gel
ltration chromatography results indicate that hirtin is a monomeric protein (Fig. 1b) as it exhibit single band indicating high
purity. Literature survey revealed that nearly all plant serine
proteases are glycoprotein. The difference in molecular weight as
determined by reducing PAGE and gel ltration chromatography
may be either due to the destruction of glycosylated moiety during
boiling before loading protein sample onto SDS-PAGE or the
destruction of weak interactions which facilitate the increased
mobility under reducing condition on SDS-PAGE. So, the apparent

Fig. 1. a) Purity and molecular mass determination of hirtin under reducing condition
on 12% SDS-PAGE: Lane 1, molecular mass standards; Lane 2, crude latex; Lane 3,
partially puried hirtin after Q sepharose chromatography. b) Elution prole of hirtin
on gel ltration chromatography (HiLoad 16/60 Superdex 200 gel ltration column) on
AKTA-Prime, Insert: SDS-PAGE of puried hirtin after gel ltration chromatography: L1,
puried hirtin, L2, molecular mass standards.

molecular mass of hirtin, used in different biochemical calculations

was assumed to be 34 kDa. The molecular mass of the known serine
proteases from plant origin vary from 19 to 110 kDa and majority of
them are found to be in between 60 and 80 kDa. The molecular
mass of most serine proteases reported from the Euphorbiaceae
family has been found in the range 43e74 kDa [25]. Some of the
brinolytic proteases isolated from different sources represent the
molecular mass range 23e43 kDa and composed of single monomeric polypeptide like hirtin [4,6,8,13].

2.2. N-Terminal and partial internal sequencing

Table 1
Summary of purication of hirtin from Euphorbia hirta latex.
Purication
steps

Total
protein
(mg)

Total
activity
Unitsa

Specic
activity
(Units/mg/min)

Fold
purity

Activity
yield (%)

Crude latex
Q Sepharose
Superdex 200

12.0
3.3
1.36

29.17
17.6
11.2

2.43
5.33
8.24

1.00
2.19
3.39

100
60.3
38.4

a
One unit of enzyme activity was dened as the amount of enzyme which gives
rise to an increase in unit absorbance at 253 nm/min of BAEE digestion in described
assay conditions.

The puried hirtin was electrophoresed and electroblotted onto

a PVDF membrane for the N-terminal sequence analysis by Edman
degradation method. The N-terminal sequence of 15 amino acids of
hirtin was found to be Y-A-V-Y-I-G-L-I-L-E-T-A-A/N-N-E. The NCBI
blast did not show any match with known serine protease amino
acid sequences in database. However, the literature survey revealed
that N-terminal amino acids sequence of hirtin showed considerable identity only to that of reported N-terminal sequence of Milin
[25] isolated from the latex of Euphorbia milli (Table 2) but not to
other plants proteases reported from Euphorbiaceae family as Miliin

106

G.K. Patel et al. / Plant Physiology and Biochemistry 52 (2012) 104e111

Table 2
Comparison of N-terminal amino acids of hirtin with other plant serine protease.
Protease/plant sources

N-terminal sequences

Reference

Hirtin (Euphorbia hirta)

Milin (Euphorbia milii)
Eumiliin (Euphorbia milii)
Proteases B (E.supina)

YAVYIGLILETAA/NNE
DVSYVGLILETD
AFLLQIIVTPPN
TTRTPNFLGLVD

Present study
Subhash et al. (2006)
Fonseca et al. (2010)
Arima et al. (2000)

peptide (R.F-I-W-L-L-T-V-P-R.A) was obtained with higher MASCOT

score (52) than the signicant (48) but it did not show any sequence
similarity with known plant proteases. The N-terminal and partial
internal sequencing results revealed that hirtin is a unique
protease.

2.3. Proteolytic assay and inhibition study of hirtin

[26] and Eumilin [14] and other plant families [22,24]. It is to be

noted that unlike hirtin which is a 34 kDa protein possessing
brinolytic activity, Milin is a 51 kDa protein and does not possess
brinolytic activity. Furthermore, the N-terminal amino acid
sequence of the hirtin is completely different from reported brinolytic enzymes isolated from different sources indicating that it is
a unique protease [4,5,13,27]. In partial internal sequencing, one

The hydrolytic activity of hirtin towards the azocasein, gelatine,

human brinogen, brin clot and other synthetic substrates was
determined. The enzyme efciently hydrolysed the azocasein as
measured spectrophotometrically. In the gelatinolytic assay, the
incorporated gelatine in the PAGE gel showed a clear zone in a blue
background of gelatine stained with CBB R-250 indicating the
proteolytic activity of hirtin. The gelatinolytic activity was performed by incubating the semi-native gel of hirtin after removal of

Fig. 2. a) In-gel gelatinolytic (zymography) assay of hirtin at pH 5.0, 7.0 and 8.0. b) Fibrinogenolytic activity of hirtin analysed on 12% SDS-PAGE under reducing conditions: Lane 1,
molecular mass standards; Lane 2, pure hirtin; Lane 3 and Lane 11, pure brinogen incubated for 120 min without hirtin, Lane 4e10, incubation of brinogen (150 mg) with hirtin
(2.0 mg) at 37
C for 5, 10, 20, 30, 60, 90 and 120; Lane 12e14, incubation of brinogen (150 mg) with hirtin (5 mg) at 37
C for 60, 90 and 120 min in 50 mM HEPES buffer pH 7.2. c)
Fibirinolytic activity of hirtin analysed on 12% SDS-PAGE under reducing conditions: Lane 1, brin clot incubated for 120 min without hirtin; Lane 2e6 brin clot incubated with
hirtin (2.0 mg) at 37
C for 10, 20, 30, 45 and 60 min in 50 mM HEPES buffer pH 7.2.

G.K. Patel et al. / Plant Physiology and Biochemistry 52 (2012) 104e111

SDS at different pH (5, 7 and 8). The in-gel gelatinolytic activity

(zymography) was found to be highest at neutral pH 7 (Fig. 2a). The
activity gels (zymography) of hirtin at different pH clearly
demonstrate its activity at a wide pH range.
The time dependent incubation of hirtin (2 mg) with human
brinogen showed that the enzyme hydrolysed Aa chain very
efciently within 5 min of incubation while the hydrolysis of Bb and
g chain is slower. The hydrolysis of Aa chain of hirtin is similar to
the Synadenium grantii latex protein [4] and Eumilin [14] which
hydrolyse the Aa chain within 5 min of incubation and more active
as compared to the other brinolytic enzyme like Pergularain e I
which could not hydrolyse Aa chain completely even after 30 min
of incubation [6] and Korean Cordyceps militaris brinolytic enzyme
which takes 60 min for hydrolysing Aa chain [14]. The hydrolysis of
Bb chain of brinogen by hirtin (2 mg) starts at 30 min of incubation
while g chain was not completely hydrolysed even after 120 min
incubation and seems to be resistant for hydrolysis. When 5 mg
hirtin was incubated with brinogen, the Bb chains almost hydrolysed in 120 min at 37
C and g showed partial hydrolysis (Fig. 2b).
In case of Eumilin and Pergularain e I, hydrolytic activity was not
observed towards the Bb and g chains even at 120 min of incubation. These results clearly demonstrate that hirtin is more effective
than Eumilin and Pergularain e I. The hydrolytic pattern of human
brinogen by hirtin is similar to that of human thrombin which is
a specic brinogenolytic enzyme and hydrolyses Aa and Bb
subunits to release the brinopeptide A and B to form brin clot
during blood coagulation [28]. The efciency of hirtin is also
comparable to other brinogenolytic and brinoltic enzymes isolated from different organisms like metalloprotease from the Rattle
snake [10]. In contrast, the snake venom protease isolated from
Taiwan habu hydrolysed the Bb chain more efciently than Aa
chain while the g remains intact [29] (Hung et al., 2001).
To evaluate the brinolytic activity, partially cross-linked brin
clot was formed using human thrombin which was further incubated with the hirtin (2 mg) for different time duration and the
hydrolytic pattern were visualised on 12% SDS-PAGE under
reducing condition (Fig. 2c). The result showed that hydrolysis of
a chain starts at 10 min of incubation and is completely hydrolysed
in time dependent manner while b and g chains are partially
hydrolysed at 60 min incubation at 37
C.
To monitor the thrombin-like activity, the puried protein (5 mg)
protein was mixed with 100 ml of citrated human plasma solution
(50 mg mL1) along with a positive control containing pure human
thrombin (1.5 U) in 100 mL of citrated human plasma solution and
negative control only contains buffer with citrated human plasma.
The positive control readily form the brin clot after addition of
thrombin while the hirtin takes more time (10 min) to form
a visible clot.
The evaluation of cleavage specicity of the hirtin towards the
various synthetic amide and ester substrates were also investigated
(Table 3). The amidolytic activity of the hirtin was investigated by
using various chromogenic substrates. The enzyme showed the
remarkable hydrolytic activity towards amide substrate p-tosGPRNA, specic for the thrombin protease but did not show
signicant activity towards other amide substrates as AAPF and
BAPNA. For the esterase activity, BAEE and BTEE were used. It was
found that hirtin efciently hydrolysed the only BAEE. These results
demonstrated that hirtin possess both esterase and amidase
activities. The enzyme exhibited extremely low or no hydrolytic
activity towards the AAV, AAPL and AAA when incubated for the
30 min at 37
C.
In order to explore the class of hirtin protease, the inhibition
study was carried out using different class of inhibitor such as
PMSF, chymostatin, leupeptin, IAA, E64, HgCl2, EDTA, and pestatin
A. The enzyme activity was signicantly inhibited by only PMSF

107

Table 3
Activity of hirtin towards different substrates.
Substrates

Characteristic

Hirtin
activity

N-p-tosyl-Gly-Pro-Arg p-nitroanilide
N-benzoyl-L-arginine-p-nitroanilide
N-suc-Ala-Ala-Pro-Phe-p-nitroanilide

Thrombin
Cathepsin H/trypsin
Cathepsin
G/chymotrypsin
Elastase
Elastase
Elastase
Chymotrypsin
Chymotrypsin
Trypsin
General
Thrombin
General

Yes
Very less
No

N-suc-Ala-Ala-Pro-Leu-p-nitroanilide
N-suc-Ala-Ala-Ala-p-nitroanilide
N-suc-Ala-Ala-Val-p-nitroanilide
N-benzoyl-L-tyrosine p-nitroanilide
N-benzoyl-L-tyrosine ethyl ester
N-benzoyl-L-arginine ethyl ester
Gelatin
Human brinogen
Azocasein

No
No
No
No
Yes
Yes
Yes
Yes
Yes

(73%) and AEBSF (85%) and not any other inhibitor (Table 4). The
leupeptin did not show inhibition of hirtin activity which indicates
that it is not a plasmin-like protease. These results demonstrated
that hirtin is a thrombin-like serine protease. It can have applications in food industry and can be developed as potential therapeutic agent for treating thrombotic disorders.

2.4. Kinetic study

The kinetic parameters were determined by using the BAEE and
p-tos-GRPNA as substrate. The apparent MichaeliseMenten
constant (Km) and maximum reaction velocity (Vmax) for the BAEE
at 37
C was calculated from the LineweavereBurk plot and found
to be 0.575 mM and 1.87  107 M s1 respectively. The turnover
number (Kcat) and catalytic efciency (Kcat/Km) were founded to be
5.3 s1 and 9.2  103 M1 s1 respectively for BAEE. The apparent
Km and Vmax for the p-tos-GRPNA were found to be 0.431 and
3.8  108 M s1 respectively and Kcat and Kcat/Km for the p-tosGRPNA were calculated to be 0.646 s1 and 1.5  103 M1 s1.

2.5. Effect of pH and temperature on enzyme activity

The effect of pH on the enzyme activity was monitored from pH
3e11. Hirtin was found to be active at broad pH range from 3 to 9
with the pH optima 7.2 (Fig. 3a). The relative activity at pH 5 and 8
were 51% and 80% respectively and reduced drastically above pH 9.
The pH optima and activity in broad pH range of hirtin is similar to
other brinolytic enzymes of latex glycoprotein [4] Korean
C. militaris [13], Neanthes japonica brinase [8], cucumicin-like
protease [30] whereas some of the brinolytic enzyme show
their optimum activity either at acidic [31] or basic pH [32]. The
report showed that the pH optima of euphorbain proteases are in
pH range of 6e8 [16].

Table 4
Effect of Inhibitors on hirtin activity.
Inhibitor type

Inhibitor name

Inhibitor conc mM

Residual activity %

Control
Serine protease

No inhibitor
PMSF
AEBSF
Leupeptin
IAA
E64
HgCl2
EDTA
Pepstatin A

e
1
1
1
1
1
1
1
1

100
27
15
98
98.0
97.0
58.0
68.5
98.0

Serine/cysteine
Cysteine protease

Metalloprotease
Aspartic protease

108

G.K. Patel et al. / Plant Physiology and Biochemistry 52 (2012) 104e111

100
2
-1
3
MRE 10 (deg cm dmol )

Relative acyivity (%)

80
60
40
20
0
2

10

12

6
4
2
0
-2
-4
-6
190

Relative activity (%)

220

230

240

Fig. 4. CD study of hirtin at pH 7.2 at 25
C.

100

2.6. Effect of metal ions, reducing agents, organic solvent,

detergents on enzyme activity

80
60
40
20
0
20

30

40

50

60

70

80

90

Temperature C
Fig. 3. a) pH optima of hirtin. b) Temperature optima of hirtin.

For the pH stability study, the residual activity of the preincubated enzyme at different pH (3e11) was measured using ptos-GPRNA as a substrate at 37
C in 50 mM HEPES buffer pH 7.2. It
was found that enzyme is completely stable between pH 4e8. The
activity at broad pH range and its pH stability conrms its potential
use in biotechnological applications. The puried hirtin was active
from temperature ranging from 30 to 70
C. The temperature
optima of hirtin was found to be 50
C (Fig. 3b) and stable for
40 min without any loss in enzyme activity while 35% activity
remains when incubated for 60 min at 50
C.
Table 5
Effect of metal ions on hirtin activity.
Metal ions

Concentration (mM)

Relative activity (%)

Control
NaCl
CaCl2

H2O
1
1
2
1
1
1
1
1
1
1

100
102
150
200
105
95
87.4
86
91
52.5
83

MgCl2
MnCl2
CuSO4
BaCl2
NiSO4
HgCl2
CoCl2

210

Wavelength (nm)

pH

200

The effect of metals ions on hirtin activity was measured in

presence of various metal salts (Table 5). The metal ions such as
Ca2 and Mg2 were found to increase the enzyme activity. The
CaCl2 at 2 mM concentration increased the hirtin activity by 2 fold
and at 5, 10 and 20 mM it remains up to 1 fold. The results indicate
that hirtin activity is not absolutely dependent on metal ions and
could be advantageous to microbial proteases which are dependent
on Ca2 for their activity. Moreover, some metal ions Co2, Hg2
and Ba2 have negative effect on the enzyme activity. Similar
observation was also reported from Dubiumin [24]. The enhanced
activity of hirtin in presence of Ca2 is similar to the most studied
proteases which interfere in blood coagulation are either serine or
metalloprotease and requires Ca2 for the activity [11].
It was found that the organic solvents as methanol, ethanol, and
DMF at 5 and 10% concentration did not affect the enzyme activity.
The DMSO did not affect enzyme activity at 5% but reduce 50%
activity if present at 10%. The detergents as SDS, Triton X-100, and
Tween-80 slightly increase the enzyme activity when present at 1%
but at 5% concentration the activity reduce by 10e15%. The SDS
increases the activity even at 5% concentration by 50%. The DTT, bmecaptoethanol, Urea did not affect the enzyme activity at 1 and 5%
concentration incubated for the 20 min. The 1% HCl solution has no
effect on activity but reduced 50% when present at 5% which might
be due to alteration of the pH of the reaction mixture.
2.7. Circular dichroism study
Far-UV CD spectra (190e240 nm) were analysed by Dichroweb
programme using CDSSTR method which showed less content of
the a-helix (13%) and high (34%) content of b-sheets at pH 7.0
(Fig. 4) which revealed that hirtin is a, b protein. The results are
similar to other reported brinogenolytic proteases which showed
less content of a-helix (7%) and high (48%) of b-sheets [4].
3. Conclusion
In conclusions, a novel thrombin-like serine protease (hirtin)
having brinolytic activity was puried and characterized from the
latex of E. hirta. The enzyme efciently hydrolyzed brinogen and
brin clot. Inhibition studies and high specicity to chromogenic

G.K. Patel et al. / Plant Physiology and Biochemistry 52 (2012) 104e111

substrate for thrombin clearly indicate that hirtin is a thrombin-like

serine protease. The enzyme exhibited stability over a wide range of
pH and temperature. The results suggest that hirtin can be developed as potential therapeutic agent for thrombotic disorders and in
other biotechnological applications. The actual biological role of the
hirtin, a major protein in the E. hirta latex needed to be further
investigated.
4. Experimental
4.1. Materials
The latex of E. hirta was collected locally. Q sepharose, casein,
human brinogen fraction I: type I, citrated human plasma,
thrombin, BSA, azocasein, gelatin, haemoglobin, skimmed milk, 2[4-(2hydroxyethyl)1-piperazinyl)] ethane sulphonic acid (HEPES),
3-cyclohexylamino-1propan-sulphonic acid (CAPS), synthetic
substrates like N-benzoyl-L-arginine ethyl ester (BAEE), N-benzoylL-tyrosine ethyl ester (BTEE), N-benzoyl-L-arginine-p-nitroanilide
(BAPNA), N-benzoyl-L-tyrosine p-nitroanilide (BTNA), N-p-tosylGly-Pro-Arg p-nitroanilide (p-tos-GPRNA), N-succinyl-Ala-Ala-ProPhe-p-nitroanilide (AAPF), N-succinyl-Ala-Ala-Pro-Leu-p-nitroanilide (AAPL), N-succinyl-Ala-Ala-Ala-p-nitroanilide (AAA), Nsuccinyl-Ala-Ala-Val-p-nitroanilide (AAV), Inhibitors like Phenylmethylsulfonyluoride (PMSF), 4-(2-Aminoethyl)benzenesulfonyl
uoride hydrochloride (AEBSF), leupeptin, EDTA, HgCl2, 1-transepoxysuccinylleucylamide(4-guanidino)butane-N-[N-(L-3-transcarboxyirane-2-carbonyl)-L-leucyl] agimatine (E64), gel regents as
SDS, acrylamide, N,N-methylene bisacrylamide, b-mercaptoethanol, coomassie brilliant blue R-250 (CBB R-250) and blotting
chemicals were purchased from Sigma Aldrich (St Louis, MO,USA).
Protein molecular weight standards and gel ltration standards
were purchased from Bio-Rad, USA. All other chemicals and
solvents trichloroacetic acid (TCA), dimethylsulfoxide (DMSO),
dimethylformamide (DMF), methanol and ethanol used were of
analytical grade.
4.2. Collection of latex and purication of hirtin
The latex of E. hirta is mainly present in the apical part of plant.
Latex was collected in 50 mM phosphate buffer pH 7.2 in fractions
of 1 mL on ice by breaking the apical tender part of plant. The
collected samples were stored at 20
C for overnight. The frozen
latex was thawed and centrifuged (Allegra X-22R, Bakeman) at
20000 g for 30 min at 4
C. The cleared supernatant was collected
and ltered through 0.45 mm syringe lter to remove all trace of
precipitated gum and the ltrate was used as crude latex extract for
protein purication. All the purication steps were performed at
4
C in a cooling cabinet. The crude latex extract from above step
was loaded on Q sepharose fast ow column (1.5  20 cm Econocolumn, Bio-Rad), pre-equilibrated with 50 mM phosphate buffer
pH 7.2. Column was washed extensively to remove the unbounded
proteins and other impurities. The bounded proteins were eluted at
a ow rate of 1 mL min1 with a linear gradient of NaCl (0e0.5 M)
in 50 mM phosphate buffer pH 7.2. Fractions having proteolytic
activity were pooled, dialysed against the 50 mM phosphate buffer
pH 7.2, concentrated and loaded onto a pre-equilibrated HiLoad 16/
60 Superdex 200 gel ltration column (16  600 mm, GE Healthcare). The protein was eluted with the same buffer at a ow rate of
0.5 mL min1 and the fraction having protease activity was pooled
for further analysis. The initial calibration of column to demonstrates the good resolution in the separation of standard proteins
and relative molecular weight of hirtin was determined by using
the gel ltration standards (Bio-Rad) as 670 kDa, thyroglobulin;
158 kDa,g-globulin; 44 kDa, ovalbumin; 17 kDa, myoglobin; and

109

1.35 kDa, vitamin B12. The concentration of puried protein was

estimated spectrophotometrically by Coomassie blue dye binding
method [33] using BSA as the standard protein. The presence of
protein during the purication steps were also monitored by taking
absorbance at 280 nm using UVeVis spectrophotometer (DU730,
Bakcman).
4.3. SDS-PAGE analysis
The crude extract of latex, Q sepharose eluted fractions and
puried hirtin were electrophoresed on 12% polyacrylamide gel
containing 0.1% SDS under reducing conditions. The protein bands
were visualized by 0.15% staining with CBB R-250. The relative
molecular mass of hirtin was also determined by using a low range
SDS-PAGE molecular weight standards (Bio-Rad) as97.4 kDa,
phosphorylase b; 66.2 kDa, bovine serum albumin; 45 kDa, ovalbumin; 31 kDa, carbonic anhydrase; 21.5 kDa, trypsin inhibitor and
14.4 kDa, lysozyme.
4.4. N-Terminal and partial internal sequencing
The puried hirtin was electrophoresed on 12% SDS-PAGE and
electroblotted onto a PVDF membrane (Immobilon-P, Millipore) at
350 mA using 100 mM CAPS buffer, pH 11.0 in mini trans-blot unit
(Bio-Rad) according to Matsudaria [34]. After electroblotting, the
membrane was rinsed 3e4 times with deionized water and stained
with 0.1% CBB R-250 in 1% acetic acid 50% methanol for a few
seconds. Membrane was destained with 50% methanol. Protein
spots were excised and rinsed for 10 min in deionized water and
air-dried. The N-terminal amino acid sequencing was performed by
Edman degradation method on an automated sequencer (Procise
491cLC; Applied Bio-systems) at the protein sequencing facility of
Institute of Microbial Technology (IMTECH), Chandigarh, India. A
database search for comparable peptides sequences was performed
by NCBI blastp program [35].
The partial internal sequencing of hirtin was done at proteomics
facility of The Centre for Genomic Application (TCGA), New Delhi,
India. The puried protein was electrophoresed on a 12% SDS-PAGE
and protein bands were excised and partially digested with trypsin.
The resulted peptides were subjected to two dimensional liquid
chromatography ESI-MS (Agilent 1100 series 2DnanoLC-MS) followed by reverse phase separation. The peptides get ionized in the
liquid phase in the Electrospray ionizer (Bruker Daltonics Ultraex
TOF/TOF) and enter the ion trap, get fragmented (MS/MS) and
detected. The data were analysed using MASCOT search engine.
4.5. Proteolytic assay and kinetic study
The hydrolytic activity of the hirtin towards different substrates
were monitored at 37
C using azocasein, gelatin, brinogen,
synthetic peptides as p-tos-GPRNA, BAPNA, BTNA, AAPF, AAPL,
AAA, AAV, esters as BAEE and BTEE as substrate.
4.5.1. Azocaseinolytic assay
The azocaseinolytic activity of puried hirtin was assayed using
1% azocasein as substrate in 50 mM potassium phosphate buffer pH
7.2. The reaction was performed in 500 ml assay volume containing
200 ml of 1% azocasein (w/v) and HEPES buffer pH 7.2 using 5 mg of
enzyme and the mixture was incubated for 20 min at 37
C. The
reaction was terminated by adding one volume of 20% TCA and
placed on ice for 10 min and centrifuged at 15,000 g for 10 min.
The supernatant was mixed with the equal volume of 1 M NaOH
solution and absorbance was measured at 440 nm with the reagent
blank.

110

G.K. Patel et al. / Plant Physiology and Biochemistry 52 (2012) 104e111

4.5.2. In-gel protease assay (Zymography)

In-gel protease assay was performed according to the method
described by Choi et al. [36] with slight modications using 0.1%
gelatin co-polymerised with the resolving gel. The puried enzyme
was mixed with non-reducing sample buffer and was run on 12%
SDS-PAGE without boiling. After electrophoresis, gel was washed
with 20% Triton X-100 for 1 h at 4
C and rinsed with distilled water
for three times at the interval of 30 min to remove all trace of SDS.
Now gels were immersed in developing buffers at pH 5.0, 7.0 and
8.0 supplemented with 2 mM CaCl2, 100 mM NaCl and incubated
for 12 h at 37
C for the protease activity. The reaction was terminated by adding ice cold 20% TCA for 20 min at room temperature.
Gel was stained with 0.15% CBB R-250 in water:methanol:acetic
acid (50:40:10) and destained with same solution without dye to
visualize the clear hydrolytic zone.
4.5.3. Fibrinogenolytic and brinolytic activity
The Fibrinogenolytic activity assay was performed as
described by Rajesh et al. [4] with slight modication. The reaction mixture contained 150 mg of brinogen, 2 mM CaCl2 in
50 mM HEPES buffer pH 7.2, was incubated with 2.0 mg of enzyme
at 37
C for 5, 10, 20, 30, 45, 60 and 120 min and also by using
5.0 mg of enzyme for 30 60 and 90 min in different tubes. A
control reaction mix was also incubated for 120 min at 37
C in
absence of hirtin. The reaction was terminated by boiling after
addition of 4X reducing sample buffer (0.25 M TriseHCl pH 6.8,
5% b-mercaptoethanol, 8% SDS, 40% glycerol and 0.2% bromophenol blue). The hydrolyzed products were analysed on a 12%
SDS-PAGE and protein pattern was visualized by staining with
0.15% CBB R-250.
For the brin clot degradation activity, partially cross-linked
brin clot was formed using 150 mg of brinogen in 50 mM
HEPES buffer pH 7.2 supplemented with 2 mM CaCl2 in presence of
1.5 U of human thrombin after incubation for 20 min at 37
C.
Partially cross-linked clot was washed with HEPES buffer pH 7.0
and incubated with 2 mg hirtin in 50 mM HEPES buffer pH 7.0
containing 2 mM CaCl2 at 37
C for 10, 20, 30, 45 and 60 min in
different tubes. The reaction was terminated and analysed on 12%
reducing SDS-PAGE as described above.

4.5.6. Kinetic study

The kinetic parameters were calculated from the initial rate of
enzymatic hydrolysis of puried hirtin using BAEE and p-tosGPRNA as substrate. The enzymatic activity towards BAEE and ptos-GPRNA were assayed using 2.0 mg hirtin protease by continuous
and discontinuous method respectively. To determine the apparent
MichaeliseMenten constant (Km), the increasing BAEE concentration (0.05e1.5 mM) was incubated with the 2.0 mg of enzyme in
1 mL reaction volume and the initial rate of hydrolysis was determined by absorbance change per min at 253 nm by continuous
method. For p-tos-GPRNA, the 500 ml reaction mixture containing
a substrate range (0.01e2.0 mM) were incubated with enzyme at
37
C for different times. After incubation, the reactions were
terminated by addition of 20% acetic acid and absorbance was taken
at 410 nm. Each assay was carried out in triplicate. The kinetic
parameters were calculated from LineweavereBurk plot.
4.6. Effect of pH and temperature on enzyme activity
To determine the optimum pH of hirtin, enzyme activity was
assayed at different pH (3e11), using p-tos-GPRNA as a substrate in
different buffers as citrate phosphate (3e8), TriseHCl (8e9) and
glycine-NaOH (10e12). For the pH stability study puried hirtin
(5 mg) was incubated for 1 h at 37
C in different buffers of pH range
(2e11) at the interval of 1 pH unit and the residual activity was
determined by further incubation for 10 min at 37
C at pH 7.2 as
described earlier.
To determine the optimum temperature, 5 mg of puried hirtin
enzyme was used in a 500 ml of reaction mixture containing 1 mM
p-tos-GPRNA in 50 mM HEPES buffer pH 7.2 with 2 mM CaCl2. The
reaction mixture was incubated for 10 min at temperature ranges
from 25 to 90
C at 5
C increment and the hydrolytic activity was
determined as described above. The stability of enzyme at optimum
temperature was determined by pre-incubating the enzyme at
temperature optima (50
C) for 0e90 min with 10 min increments
and the residual activity was determined using p-tos-GPRNA as
a substrate at 37
C. The non heated enzyme was considered as
positive control and assumed to have 100% activity.
4.7. Protease inhibition assay

4.5.4. Thrombin-like activity

The lyophilised citrated human plasma (sigma) was dissolved in
sterile, nuclease and protease free distilled water. The 100 ml of
human plasma (50 mg mL1) was incubated in different tubes with
different amount of hirtin (2 and 5 mg) in 50 mM HEPES buffer pH
7.2 having 2 mM CaCl2. The time for visible clot formation was
recorded.
4.5.5. Activity towards synthetic substrates
The enzymatic hydrolysis of different synthetic substrates
peptidyl-pNA (peptidyl p-nitroanilide) by the puried hirtin was
studied by spectrophotometric method. The synthetic substrates
used were BAPNA, BTNA, p-tos-GPRNA, AAPF, AAPL, AAA, and AAV.
The 500 ml reaction mixture containing 1 mM substrates, 2 mM
CaCl2 in 50 mM HEPES buffer pH 7.2 was initially equilibrated at
37
C for 5 min and further incubated for 10 min after addition of
5 mg of enzyme. The reaction was terminated by addition of 500 ml
of 20% acetic acid. The rate of hydrolytic activity was determined by
measuring the released p-nitroaniline at 410 nm using the molar
extinction coefcient 410 8480 M1 cm1.
The hydrolytic activity towards the ester substrates, N-benzoylL-arginine ethyl ester (BAEE) and N-benzoyl-L-tyrosine ethyl ester
(BTEE) were examined spectrophotometrically at 253 nm
(253 1150 M1 cm1) and 259 nm (259 964 M1 cm1)
respectively.

The protease inhibition assay was done using the PMSF, AEBSF,
leupeptin, E64, HgCl2, EDTA, pestatin A to determine the class of
protease. The 1 mM inhibitor solution was incubated with 5 mg of
the puried hirtin in 50 mM phosphate buffer pH 7.2 and the
residual activity was determined using BAEE as substrate as
described above. The control assay was done in absence of the
inhibitor and the activity was taken as 100%.
4.8. Effect of metal ions on enzyme activity
The effect of metal ions on the enzyme activity was determined
using different salts in 50 mM HEPES buffer pH 7.2. The salt of Na,
Ca2, Mg2, Mn2, Hg2, Ba2 and Co2 were added as chlorides,
Cu2 and Ni2 were added as sulphates. The entire assay was done
in triplicate using p-tos-GPRNA as a substrate and average was
taken as data point.
4.9. Effect of denaturants, organic solvents and detergents on
enzyme activity
The effect of different surfactants like SDS, Tween-20, Tween-80,
Triton X-100, Urea (1 and 5%), organic solvents methanol, ethanol,
DMSO, DMF (1 and 10%) and reducing agents DTT, b-mercaptoethanol (1and 5%) on enzyme activity was studied by pre-incubating

G.K. Patel et al. / Plant Physiology and Biochemistry 52 (2012) 104e111

the puried hirtin at 37
C for 30 min with these effectors. The

relative activity was determined in 50 mM HEPES buffer pH 7.2
supplemented with 2 mM CaCl2 and 1 mM p-tos-GPRNA as
a substrate. The enzyme activity was assumed as 100% in absence of
any additive.
4.10. Circular dichroism study
Circular Dichroism study was performed on Chirascan CD
spectrometer (Applied Photophysics, UK). Far-UV CD spectra
(180e260) were recorded in a 1 mm quartz cell at bandwidth 1 nm
and time per point 0.5 s with three repeats. The CD study was done
using 0.2 mg mL1 puried protein in 20 mM Citrate phosphate
buffer pH 7.0 at 25
C. Three consecutive scans were accumulated.
The average base line spectrum of buffer blank was subtracted from
the average protein sample spectrum and analysed (190e240 nm)
by online DichroWeb programme using CDSSTR method [37]. The
results of the CD measurements were expressed as mean residue
ellipticity (MRE) in deg cm2 dmol1.
Acknowledgements
CD studies were performed at NMR facility at Institute Instrumentation Centre (IIC), IIT Roorkee. G. K. Patel and A. A. Kawale
gratefully acknowledge the nancial support from Ministry of
Human Resource Development (MHRD) and Department of
Biotechnology (DBT), Government of India, respectively.
References
[1] H.R. Lijnen, Matrix metalloproteinases and cellular brinolytic activity,
Biochemistry (Moscow) 67 (2002) 92e98.
[2] K.L. Berkner, Blood Clotting: General Pathway, Encyclopedia of Life Sciences,
John Wiley & Sons, Ltd, 2001. doi:10.1038/npg.els.0001408.
[3] I.D. Walker, J.F. Davidson, J.M. Thomson, Blood Coagulation and Hemostasis,
Churchill Livingstone, Edinburgh, 1985, 208e263.
[4] R. Rajesh, A. Natarajua, C.D.R. Gowda, B.M. Frey, F.J. Frey, B.S. Vishwanath,
Purication and characterization of a 34-kDa, heat stable glycoprotein from
Synadenium grantii latex: action on human brinogen and brin clot, Biochimie 88 (2006) 1313e1322.
[5] H.V. Shivaprasad, R. Rajesh, B.L. Nanda, K.K. Dharmappa, B.S. Vishwanath,
Thrombin like activity of Asclepias curassavica L. latex: action of cysteine
proteases, J. Ethnopharmacol. 123 (2009) 106e109.
[6] H.V. Shivaprasad, R. Rajaiah, B.M. Frey, F.J. Frey, B.S. Vishwanath, Pergularaine
I- a plant cysteine protease with thrombin-like activity from Pergularia
extensa latex, Thrombos. Res. 125 (2010) e100ee105.
[7] Z. Deng, S. Wang, Q. Li, X. Ji, L. Zhang, M. Hong, Purication and characterization of a novel brinolytic enzyme from the polychaete, Neanthes japonica
(Iznka), Bioresour. Technol. 101 (2010) 1954e1960.
[8] S. Wang, Z. Deng, Q. Li, X. Ge, Q. Bo, J. Liu, J. Cui, X. Jiang, J. Liu, L. Zhang, M. Hong,
A novel alkaline serine protease with brinolytic activity from the polychaete,
Neanthes japonica, Comp. Biochem. Physiol. Part B. 159 (2011) 18e25.
[9] B.S. Hahn, S.Y. Cho, M.Y. Ahn, Y.S. Kim, Purication and characterization of
a plasmin-like protease from Tenodera sinensis (Chinese mantis), Insect Biochem. Mol. Biol. 3 (2001) 573e581.
[10] S. Chiou, C. Hung, K. Huang, Characterization of a protease with a- and bbrinogenase activity from the western diamondback rattlesnake, Crotalus
atrox, Biochem. Biophys. Res. Commun. 187 (1992) 389e396.
[11] S. Swenson, F.S. Markland Jr., Snake venom brin(ogen)olytic enzymes, Toxicon 45 (2005) 1021e1039.
[12] V. Deepak, K. Kalishwaralal, S. Ramkumarpandian, S.V. Babu,
S.R. Senthikumar, G. Sangiliyandi, Optimization of media composition for
Nattokinase production by Bacillus subtilis using response surface methodology, Bioresour. Technol. 99 (2008) 8170e8174.

111

[13] D. Choi, W.S. Cha, N. Park, H.W. Kimb, J.H. Lee, J.S. Park, S.S. Park, Purication
and characterization of a novel brinolytic enzyme from fruiting bodies of
Korean Cordyceps militaris, Bioresour. Technol. 102 (2011) 3279e3285.
[14] K.C. Fonseca, N.C.G. Morais, M.R. Queiroz, M.C. Silva, M.S. Gomes, J.O. Costa,
C.C.N. Mamede, F.S. Torres, N.P. Silva, M.E. Beletti, H.A.N. Canabrava,
F. Oliveira, Purication and biochemical characterization of Eumiliin from
Euphorbia milii var. hislopii latex, Phytochemistry 71 (2010) 708e715.
[15] N.O. Cafni, L.M.I. Lopez, C.L. Natalucci, N.S. Priolo, Proteases of higher plants:
General features, physiological roles and applications, Acta Farm. Bonaerense
7 (1988) 195e213.
[16] K.R. Lynn, N.A. Clevette-Radford, Proteases of Euphorbiaceae, Phytochemistry
27 (1988) 45e50.
[17] K.R. Kiritikar, D.J. Basu, in: K.S. Mhaskar, E. Blatter, J.F. Caius (Eds.), Illustrated
Indian Medicinal Plants, 9, Satguru Publications, New Delhi, 2000, pp.
3031e3033.
[18] S.B. Patil, N.S. Naikwade, C.S. Magdum, Review on phytochemistry and
pharmacological aspects of Euphorbia hirta Linn, J. Pharmacol. Res. Health Care
1 (2009) 113e133.
[19] M.C. Lanhers, J. Fleurentin, P. Dorfman, F. Mortier, J.M. Pelt, Analgesic, antipyretic and anti-inammatory properties of Euphorbia hirta, Planta Med. 57
(1991) 225e231.
[20] J.F. Akinrinmade, O.A. Oyeleye, Antimicrobial efcacy and tissue reaction of
Euphorbia hirta ethanolic extract on canine wounds, Afr. J. Biotechnol. 9
(2010) 5028e5031.
[21] P.B. Johnson, E.M. Abdurahman, E.A. Tiam, I. Abdu-Aguye, I.M. Hussaini,
Euphorbia hirta leaf extracts increase urine output and electrolytes in rats,
J. Ethnopharmacol. 65 (1999) 63e69.
[22] M. Pande, V.K. Dubey, S.C. Yadav, M.V. Jagannadham, A novel serine protease
cryptolepain from Cryptolepis buchanani: purication and biochemical characterization, J. Agric. Food Chem. 54 (2006) 10141e10150.
[23] S.C. Yadav, M. Pande, M.V. Jagannadham, Highly stable glycosylated serine
protease from the medicinal plant Euphorbia milii, Phytochemistry 67 (2006)
1414e1426.
[24] I.A.M. Ahmed, I. Morishima, E.E. Babiker, N. Mori, Dubiumin, a chymotrypsinlike serine protease from the seeds of Solanum dubium Fresen, Phytochemistry
70 (2009) 483e491.
[25] C.M. Antao, F.X. Malcata, Plant serine proteases: biochemical, physiological
and molecular features, Plant Physiol. Biochem. 43 (2005) 637e650.
[26] L.P. Moro, M.T. Murakami, H. Cabral, A. Vidotto, E.H. Tajara, R.K. Arni,
L. Juliano, G.O. Bonilla-Rodriguez, Purication, biochemical and functional
characterization of miliin, a new thiol-dependent serine protease isolated
from the latex of Euphorbia milii, Protein Pept. Lett. 15 (2008) 724e730.
[27] C.T. Chang, M.H. Fan, F.C. Kuo, H.Y. Sung, Potent brinolytic enzyme from
a mutant of Bacillus subtilis, J. Agric. Food Chem. 48 (2000) 3210e3216.
[28] Y. Komori, H. Sugihara, A.T. Tu, Specicity of haemorrhagic proteinase from
Crotalus atrox (Western diamond back rattle snake) venom, Biochim. Biophys.
Acta 829 (1985) 127e130.
[29] Chin-Chun Hung, Shyh-Horng Chiou, Fibrinogenolytic proteases isolated from
the snake venom of Taiwan habu: serine proteases with kallikrein-like and
angiotensin-degrading activities, Biochem. Biophys. Res. Commun. 281 (2001)
1012e1018.
[30] K. Arima, T. Uchikoba, H. Yonezawa, M. Shimada, M. Kaneda, Cucumisin-like
protease from the latex of Euphorbia supine, Phytochemistry 53 (2000)
639e644.
[31] L. Cui, M.S. Dong, X.H. Chen, M. Ziang, X. Lv, G. Yan, A novel brinolytic
enzyme from Cordyceps militaris, a Chinese traditional medicinal mushroom,
World J. Micriobiol. Biotechnol. 24 (2008) 483e489.
[32] J.S. Kim, S. Kumar, S.E. Park, B.S. Choi, S. Kim, T.H. Nguyen, C.S. Kim, H.S. Choi,
M.K. Kim, H.S. Chun, Y. Park, S.J. Kim, A brinolytic enzyme from the medicinal
mushroom Cordyceps militaris, J. Microbiol. 44 (2006) 622e631.
[33] M.M. Bradford, A rapid sensitive method for the quantitation of microgram
quantities of protein utilizing the principle of protein-dye binding, Anal.
Biochem. 72 (1976) 248e254.
[34] P. Matsudaria, Sequence from picomole quantities of proteins electroblotted
onto polyvinylidene diuoride membranes, J. Biol. Chem. 262 (1987)
10035e10038.
[35] S.F. Altschul, W. Gish, W. Miller, E.W. Myers, D.J. Lipman, Basic local alignment
search tool, J. Mol. Biol. 215 (1990) 403e410.
[36] Nack-Shick Choi, Kab-Seog Yoon, Jin-Young Lee, Kyoung-Yoen Han, SeungHo Kim, Comparison of three substrates (casein, brin, and gelatin) in
zymographic gel, J. Biochem. Mol. Biol. 34 (2001) 531e536.
[37] L. Whitmore, B.A. Wallace, DICHROWEB: an online server for protein
secondary structure analyses from circular dichroism spectroscopic data,
Nucl. Acids Res. 32 (2004) W668eW673.