全 文 ：38　 Jou rna l of Ch inese Pha rm aceutica l Sc iences 2006, 15(1):38-44
Interactions Between Thromb in and NaturalProudcts ofCoreopsis
tinctoria Nuttt. andC istanche deserticolaMa. in
Capillary Zone E lectrophoresis
LIU Y i
1 , ZHANG Yuan2 , LIU X iao-m ing2 , LING X iao-mei1*, TU Peng-fei2 , and CUI Jing-rong3**
(1. Departmen t of Pharmaceu tical Ana lysis, S chool of Pharmaceu tical S ciences, Pek ing University, Beijing 100083 , China;
2. Departmen t of Natural Medicines, School of Pharmaceutica l Science, Pek ing Un iversity, B eijing 100083 , China;
3. S tateKey Laboratory of Natural and B iom im eticD rugs, Pek ing Un iversity, B eijing 100083 , China)
Received date:2005-10-08.
Foundat ion item:National H igh-tech Research and Developm en t
P rogram ofC hina (2004AA2Z3783).
* Correspond ing au thor:Tel86-10-82801590,
E-m ai l pku lingxm@ s ina. com ,
　　　　　　　　　　** Tel86-10-82802467,
E-mai l jrcui@ b jmu. edu. cn
Abstract:Ami 　A capillary zone e lectrophoretic method (CZE)was used to de term ine the inter-
ac tions betw een natural products and thrombin. Methods　Samples con taining natural products and
thrombin at various ratios w ere incubated at 25 ℃ and then w ere separated by CZE w ith Tris-ace tate
buffer at pH 7. 2. Each run could be accomplished w ithin 5 m in. Results　 In CZE , the peak w id th
b roadened due to the affin ity interaction betw een na tu ra l products and th rombin. Compa red w ith positive
and negative contro l, the natural products(CB-1 , CB-2) from Coreopsis tinctoria Nu tt.t interac ted w ith
thrombin;CB-3 from Coreopsis tinctoria Nu tt.t andHC-1, HC-2, HC-3 from Cistanche deserticola Ma.
did not b ind to thromb in. Both qualification and quan tification charac te riza tions o f the b ind ingw e re de-
te rm ined. Conclusion　The establishedme thod is capable of sensitive and fast de term ina tion o f natural
produc ts and thrombin interactions, it can be employed as an a lte rnative method.
Key words:　capillary zone electropho resis;Coreopsis tinctoria Nutt.t ;Cistanche deserticola Ma;
in te rac tions;th rombin
CLC number:R917;R917. 7　Docum ent code:A　Article ID:1003-1057(2006)1-038-07
In troduction
Coreopsis tinctoria Nutt.t is w idely d istribu ted in
America, South A frica and H aw aiian Islands and has
been p lanted in C hina fo r a long time, and m any of
them are used as a the rapeu tica l agen t for inflamma tion
and d iarrhea. CistanchedeserticolaMa. is a short para-
sitic p lant native in the northw est o fCh ina. The who le
dried plant (w ithout flow ers) is know n as Herba
Cistanch is used in o rien talmedic ine fo r the trea tmen t of
kidney pain, gynaeco logica l diseases, and in testinal
infections
[ 1 -3]
. In folk medic ine, tho se tw o p lants are
employed as preventives fo r cardiovascular diseases.
Three ingred ients we re iso la ted from each o f the tw o
p lants, namely, iso-okaning lucopyrano side (CB-1),
okan inglucopy ranoside (CB-2), and querce tagitrin
(CB-3) from Coreopsis tinctoria Nutt.t , and acteoside
(HC-1), tubu lo side (HC-2), and genipo side (HC-
3) from Cistanche deserticola M a. The chem ical struc-
trures o f these compounds are shown in Figure 1.
Thrombin is a serine protease in the inactiva tion o f
the b lood coagulation system. It initia tes b lood clotting
by cleav ing soluble fibrinogen and a lso ac tivates o ther
procoagu lant enzymes including facto rs V , V III, X I
and XIII, and the an ti-coagu lant enzyme protein C.
Thrombin also has ma rked e ffec ts on a variety o f cells
including p latele ts, endothe lial and smoo th muscle
ce lls, which seem to be transm itted via specific throm-
bin recepto rs ca lled protease-ac tivated recep to rs[ 4] .
L IU Y i, et a l:In te rac tions Be tw een Thrombin and Na tu ra l P roudcts of Co reopsis tinc to riaNu tt.t
＆Cistanche deserticolaM a. in Capilla ry Zone E lec trophoresis 39　
Consequently, most cu rrent an tithrombotic treatment
strategies are aimed at blocking the activ ity of throm-
b in , or p reventing its genera tion[ 5] . It is an important
targe t for therapeutic inte rven tion in thrombotic d is-
ease, and a number o f strateg ies have been successfu l-
ly app lied to the developmen t of po tent thromb in inhib-
ito rs. Screening o f na tura l and synthetic compounds
based on the concep t o f ligand /receptor inte raction is
an important strategy to sea rch fo r thromb in inhibitors.
M any methods have been emp loyed to study the
in terac tions between prote ins and d iffe rent compounds,
such as nuclearm agnetic resonance (NMR), Fou rier
transfo rm infrared spectrome try (FTIR), a ffinity chro-
ma tography(AC), mass spec trometry(MS), X-ray,
fluo rescence, surface p lasmon resonance spec trometry
(SPR), po lyacry lam ide ge l e lec trophoresis (PAGE),
equ ilibrium dialysis, and capillary e lectrophoresis
(CE). Among the seme thods, NMR, X-ray, MS , f lu-
orescence, and FTIR require elaborated procedures
such as sample purification befo re ana ly sis. SPR , AC ,
PAGE , and equ ilibrium dia ly sis are time-consum ing
procedures and require a large samp le size fo r analy sis.
CE has been used in a w ide range o f binding stud ies
betw een important bio logica lly molecu les, such as pro-
tein-pro tein, prote in-DNA , pro tein-drug, p ro tein-sug-
a r, DNA-peptide, peptide-drug, peptide-peptide, an-
tibody-an tigen, and peptide-carbohydrate[ 6] . Among
the va rious CE modes
[ 7, 8] , CZE is themost appropriate
one fo r the study o fmolecular in te ractions if only sm all
amounts o f samples are availab le. In th is study, due to
its high speed, excellent reso lution, sma ll samp le size,
high reproducib ility and flexibility, CE was used to in-
vestigate the bind ing be tw een na tura l products and tar-
get proteins. CE can also be used to study the interac-
tion o f individual components in a m ix tu re as w ell as to
de term ine b ind ing pa rame te rs in a sing le step.
Expe rimenta l
Instrumenta tion
A ll expe rimen ts we re perfo rmed on a Beckman P /
ACE
TM
MDQ system (Beckman Cou lte r, Inc. , Fu ller-
ton, CA , USA)equipped w ith a photod iode a rray de-
tector and the 32 K arat
TM
softw are version 5. 0 (Beck-
man). A capillary tube (Yongnian Optical F ibre
C orp. , Hebe i, China)w ith an in te rnal d iameter o f 75
μm was used. The to tal and effective leng ths o f the
capillary w ere 60. 2 cm and 50 cm , respectively. Be-
fore use, the new capillaries we re rinsed w ith 0. 1
mol L -1 N aOH so lu tion for 20m in, and subsequently
w ith de ionized w ate r for 5 m in.
F igure 1　Structures of com pounds
40　 Journa l o f Chine se Pharm aceu tical Sciences 2006, 15(1):37-44
　　 Chem icals and reagents
A ll chem icalsw ere o f analytica l grade unless other-
w ise indicated. Human thrombin (2 000NIH units mg- 1
protein)was purchased from S igm a (S.t Louis, MO ,
USA). H epa rin and h irudin we re k indly prov ided by
Pro.f Zhang Q iang (Department of Pharmaceu tics,
Schoo l of Pharmaceutica l Science s, Peking university,
Be ijing, Ch ina). T ris base (ultrapu re) and ace tic
acid used in this study w ere from Be ijing C hem ica lRe-
agen t Fac to ry (Beijing, China). Thrombin w as d is-
solved in Tris-ace tate buffe r (30 mmo l L -1 Tris, pH
ad justed to 7. 2 w ith acetic acid) a t 1 mg mL -1
(2N IH units μL -1) as stock ing so lution. Deion ized
w a terw as prepared using aM illiporeM illiQ-Plus sys-
tem (M illipore, Bedfo rd, MA). A ll buffers and so lu-
tions used in the study w ere filtered through 0. 45 μm
membranes (Ag ilent, Germany) be fore use. Com-
pounds CB-1 , CB-2, CB-3, HC-1, HC-2, and HC-3
w ere provided by Pro.f Tu Peng-fei (Department of
N aturalM edicine s, School o fPharmaceu tica lSciences,
Peking University, Be ijing, China).
Sample preparation
To investiga te natural products-thrombin inte rac-
tions, different concentrations of th rombin w ere tested
for the forma tion of complex, which w as fo rmed by
m ixing na tura l products w ith thrombin in the running
buffer. Itw as incubated for 20m in at 25 ℃ befo re CE
analysis. A ll so lu tions w ere prepa red w ith deion ized
w a ter.
CZE of natura l products-thrombin interac-
tions
To study the in te ractions o f natural produc ts and
th rombins, the temperatures o f the cartridge and sam-
ple room were kept a t 25 ℃ and 4 ℃, respec tive ly.
Befo re each measurement, the capillary w as rinsed
w ith runn ing bu ffer (30 mmo l L -1 Tris-HA c, pH
7. 2). Samples containing them ix tures of natural prod-
ucts and th rombin w ere injected using the pressure in-
jec tionmode a t 0. 5 p. s. .i for 10 s. The app lied vo lt-
age w as 15 kV. The capillary w asw ashed be tw een runs
w ith the running buffer fo r 5 m in at 20 p. s. .i Each
concentration w as run in duplicate.
Quantitativemode l for the binding study
In the binding studies, the binding constants are
important parameters. Sca tcha rd ana ly sis is a common
w ay to linearize the binding data, as expressed in Eq.
(1):
r /C f =-Kr+nK (1)
where r is the ratio o f the concentrations of the bound
ligand (o r recep tor) to the to tal recep to r (o r ligand)
andC f is the concentration of the unbound ligand (or
recep to r). K is the appa rent b ind ing constan t and n is
the numbe r o f binding sites.
Resu lts
Comparison of coated and non-coated capillaries
Manym acrob iomo lecules are prone to be adsorbed
e lectrosta tica lly on the surface of fused-silica capillar-
ies. In genera l, this phenomenon is unfavorab le, or e-
ven fata l to the ana ly sis of macrobiomo lecu les. In our
experiments, we compa red the effects of hirud in ana ly-
sis in po lyacry lam ide-coa ted and non-coated capillar-
ies. For the non-coated capillaries, the peak appeared
symmetrical and provided h ighe r sensitiv ity. Add ition-
a lly, non-coa ted capillaries matched w ith pho todiode
array detector, thus, the structure inform ation and ab-
sorption spectrum conld be obtained simu ltaneously.
Therefo re, a non-coa ted capillary w as superio r to a
coated capillary in the binding assay.
Optmi ization of separation conditions
No sign ifican t change w as observed in the area o f
peak due to complex formation betw een 20 m in, 30
m in, 40 m in , 50 m in, and 60 m in, indica ting that e-
quilibrium was reached in so lu tion w ithin 20 m in. A ll
samples w ere, therefo re, incubated fo r at least 20 m in
be fore CE ana ly sis.
pH w as anothe r impo rtan t factors fo r eva luating
the in te ractions in capillary zone e lectrophoresis. Tris-
HA c provided the better separa tion e fficiency and reso-
lution. Thus, 30 mmo l L - 1 pH 7. 2 Tris-HA c solution
w as se lec ted as the running buffer. Unde r the em-
ployed CE conditions, thrombin w as inv isib le.
Interactions be tween hirud in /heparin and
thromb in in CZE
Thrombin inhibitors can be classified as indirect
o r direc.t Indirec t th rombin inh ib ito rs, such as heparin
and vitam in K antagon ists, b lock the gene ration and
action o f thrombin e ithe r by activating naturally occur-
L IU Y i, et a l:In te rac tions Be tw een Thrombin and Na tu ra l P roudcts of Co reopsis tinc to riaNu tt.t
＆Cistanche deserticolaM a. in Capilla ry Zone E lec trophoresis 41　
ring thrombin inhibito rs o r by inhibiting specific facto rs
in the coagu lation system tha t subsequently affect
th rombin generation o r activity. In contrast, direct
th rombin inh ib ito rs(DTIs), such as hirudin, bivaliru-
din , argatroban, andme laga tran, achieve their an tico-
agu lant effect by directly b ind ing to the thrombinmo le-
cule, thereby preventing it from in te racting w ith its
substrates
[ 9]
. So, heparin and hirudin w ere selected as
negative contro l and positive contro l in this study.
F igure 2　E lec tropherogram s of (a) hepar in-thromb in in teraction and (b) h irud in-throm bin in terac tion. H eparin /H irud in:
thrombin(V /V) (A) 10∶0;(B) 10∶3;(C) 10∶6;(D) 10∶9;(E) 10∶12;(F) 10∶15. The cond itions g iven we re a s fo llow s:
Beckm an P /ACEMDQ capillary e lec trophoresis sy stem. Injection:0. 5 p. s. .i fo r 10 s. Applied vo ltage:15 kV. C apilla ry:capillary
o f 60. 2 cm (e ffec tive leng th 50 cm) ×75 μm ID
　　Figure 2a show s the electropherog ram s for hepa rin
alone and m ixtures con taining a fixed concentra tion of
heparin and various concen trations o f th rombin. W ith
the inc rease of thrombin concentra tion, the peak he ight
of heparin w as no t changed, indicating no binding of
th rombin. F igure 2b show s the e lec tropherog ram s for
hirudin a lone and m ixtures containing a fixed concen-
tration of hirudin and various concen trations o f throm-
b in. From (A), we can see that the shape of h irudin
is irregular due to the impuritie s. W ith increasing con-
centra tions o f thromb in, the peak heigh t o f hirudin is
decreased, indicating strong binding. The interaction
betw een hirudin and thrombin leng thens the hirudin zone
and causes the peak broadening. According ly, the peak
height of hirudin decreases, bu t the peak areas do not
change, and the peak heights only present the value of
the orig ina l unbound hirudin concentrations in the m ix-
tures containing hirudin and thrombin
[ 6, 10]
.
Standard samp les con taining hirudin in the range
o f 0. 06 - 1. 00μg μL -1 , were in jected into the cap-
illary, the peak he igh ts being proportional to concen-
trations. The re lationsh ip be tw een peak heigh t and h ir-
udin concentration w as expressed as y =25 993x +
376. 1 (n = 7), w ith a co rrela tion coe fficient o f
0. 997 5. F igure 3 is the Scatchard p lo t conducted by
using the CE da ta of various samples. The binding
constant fo r hirud in w ith thrombinK was ca lculated to
be 7. 85×105 L mo l- 1 from the slope.
Interactions between natural products and
thromb in in CZE
The interactions between diffe rent natu ra l products
and th rombin a t pH 7. 2 w e re also studied. F igure 4
show s electropherog rams for CB-1 /CB-2 alone and
m ix tures containing a fixed concentra tion o f CB-1 /CB-
42　 Journa l o f Chine se Pharm aceu tical Sciences 2006, 15(1):37-44
2 and increasing concen trations of thrombin. The first
peak is considered as CB-1-th rombin complex, wh ile
the second peak is free CB-1 in F igure 4a. In Figure
4b CB-2-th rombin complex is the first peak, the later
is CB-2. The increase in the concentra tion of thrombin
in the samp le decreases the peak he ight of CB-1 /
CB-2. According to the absorp tion spectrum o f free
CB-1 /CB-2 and their complexes obta ined from the
PDA detecto r, the first peak proved to be CB-1 /CB-2-
th rombin comp lexes and the second peak as free CB-1 /
CB-2. To quantify the in te raction be tw een CB-1 /CB-2
and thromb in, the free CB-1 /CB-2 concentra tions w ere
also ca lculated from the corresponding calibration curves
and from the slopes of the Scatchard plotK were calculat-
ed to be 3. 6 ×104 L mo l-1 and 1. 13 ×105 L mo l-1.
Figure 3　Scatchard p lot for interaction between thombin and
h irud in. Experimental conditionsw ere the same as in F igure 2
F igure 4　 Electropherogram s of (a) CB-1-thrombin interact ion and (b) CB-2-thrombin interact ion. N atu ra l produc ts:
Thromb in (V /V) (A) 10∶0;(B) 10∶3;(C) 10∶6;(D) 10∶9;(E) 10∶12. The conditions g iven we re as fo llow s:Beckm an P /
ACEMDQ cap illa ry e le ctropho resis sy stem. Injection:0. 5 p. s. .i fo r 10 s. Applied voltage:15 kV. Capillary:capillary o f 60. 2 cm
(effective length 50 cm) × 75 μm ID
　　Figure 5 demonstra tes the elec tropherog rams fo rCB-
3, HC-1, HC-2, HC-3 alone, and m ixtures containing a
fixed concentration o f those products and increasing con-
centra tions of thrombin. W ith increasing thrombin con-
centra tion, the peak height and the abso rption spectrum
of CB-3, HC-1, HC-2, HC-3 we re not changed signifi-
cantly. No inte ractionsw ere obtained.
D iscussion
The presentmethod of capillary zone e lectrophoresis
was shown to be capable o f de te rm inating the natural
products and thrombin interactions. Thismethod is sensi-
tive and fast, requiring less amounts of reagents than oth-
erme thods such as SPR, AC, etc. Each run was comple-
ted w ithin 5m in and the binding constants cou ld be de-
term ined. Compared w ith positive and negative contro ls,
the natura l products(CB-1, CB-2) from Coreopsis tincto-
ria Nutt.t interacted w ith thrombin. CB-3 from Coreopsis
tinctoria Nutt.t and HC-1, HC-2, HC-3 from Cistanche
deserticolaMa. had no binding to thrombin.
L IU Y i, et a l:In te rac tions Be tw een Thrombin and Na tu ra l P roudcts of Co reopsis tinc to riaNu tt.t
＆Cistanche deserticolaM a. in Capilla ry Zone E lec trophoresis 43　
The present study show s it is possible to characterize
interactions be tween natural products and thrombin using
CZE as a new technique. An impo rtant practical aspect is
that the sample may no t be purified prior to CE separa-
tion. Therefore, a large number o f samples can be
screened by a sing le CE run for biological activity
[ 3]
.
F igure 5　Electropherogram s of(a) CB-3-thrombin interact ion, (b) HC-1-thromb in in terac tion, (c) HC-2-thrombin in ter-
act ion, and (d) HC-3-thrombin in teraction. Natura l produc ts:Thromb in(V /V) (A) 10∶0;(B) 10∶3;(C) 10∶6;(D) 10∶9;
(E) 10∶12. The conditions g iven w ere as fo llow s:Beckm an P /ACEMDQ capillary e lectropho resis sy stem. Injection:0. 5 p. s. .i fo r
10 s. App lied vo ltage:15 kV. Capillary:cap illa ry o f 60. 2 cm (effective leng th 50 cm) × 75 μm ID
Re ferences
[ 1] ChinesePharmacepoeia I [ S] . 2000. 103-104.
[ 2] L in LW , H sieh MT, Tsai FH , et al. Anti-nociceptive
and anti-inflamm atory ac tivity caused byCistanche deserti-
cola in roden ts [ J] . J E thnopharm acol, 2002, 83 (3):
177-182.
[ 3] Ouyang J, W ang XD, Zhao B, et al. E ffects o f ra re earth
e lements on the grow th o fCistanche deserticola ce lls and
the produc tion of pheny le thano id g lyco sides [ J] . J B io-
technol, 2003, 102 (2):129-134.
[ 4] LopezM , Mende K, Steinm etzer T. C loning, purifica tion
and biochem ica l characterization o f d ipeta rudin, a new
chim eric th rom bin inhibito r [ J] . J Chromatogr B, 2003,
44　 Journa l o f Chine se Pharm aceu tical Sciences 2006, 15(1):37-44
786(1-2):73-80.
[ 5] W eitz JI, C row the rM. D irec t thrombin inhibito rs [ J] .
Throm b R es, 2002, 106 (3):V275-284.
[ 6] H e XY, D ing YS, L i DZ, et al. Recent advance s in the
study of b iomo lecu la r interac tions by capillary elec tropho-
resis [ J] . E lectrophoresis, 2004, 25 (4-5):697-711.
[ 7] Sh im ura K, Kasai K. A ffinity capillary electropho re sis:a
sensitive too l for the study o f m o lecular interactions and
its use in m icrosca le ana ly sis [ J] . Anal B iochem , 1997,
251(1):1-16.
[ 8] Heegaard NHH , Kennedy RT. Identifica tion, quan tita-
tion, and cha rac te riza tion o f biom o lecules by capillary
e lec trophoretic ana lysis o f b ind ing interactions [ J] . E lec-
trophoresis, 1999, 20 (15-16):3122-3133.
[ 9] W eitz JI. A nove l approach to throm bin inh ibition [ J] .
Thromb Res, 2003, 109 (Suppl):S17-22.
[ 10] H e XY, L i DZ, L iangAY. In te rac tion between netropsin
and doub le-stranded DNA in capillary zone e lectropho re-
sis and a ffinity cap illa ry elec trophore sis [ J] . J Chrom a-
togr A , 2002, 982 (2):285-291.
毛细管区带电泳研究凝血酶与蛇目菊和肉苁蓉
提取物的相互作用
刘　一1 , 张　媛 2 , 刘晓明2 , 凌笑梅 1 , 屠鹏飞 2 , 崔景荣 3
(1.北京大学 药学院 药物分析教研室 , 北京 100083;2. 北京大学 药学院 天然药物学系 , 北京 100083;
3. 北京大学 天然药物及仿生药物国家重点实验室 , 北京 100083)
摘要:目的　采用毛细管区带电泳法研究凝血酶与蛇目菊和肉苁蓉提取物的相互作用。方法　研究了在 pH 7. 2的 T ris-
醋酸缓冲溶液中 , 凝血酶和天然提取物于 25 ℃以不同比率混合温育 20 m in后的相互作用情况 , 分离在 5 m in内完成。结果　
与阳性对照和阴性对照相比 , 蛇目菊中的提取物 CB-1, CB-2与凝血酶有相互作用 ,并计算得到了 CB-1, CB-2和凝血酶的结合
常数;蛇目菊中的提取物 CB-3和肉苁蓉中的 HC-1, HC-2, HC-3与凝血酶没有相互作用。结论　本实验采用的毛细管区带电
泳研究天然提取物与凝血酶相互作用的方法快速 、准确 、可靠。
关键词:毛细管区带电泳;蛇目菊;肉苁蓉;凝血酶;相互作用