Synthesis and evaluation of nitric oxide-releasing derivatives of 3-n-butylphthalide as anti-platelet agents

Article abstract of DOI:10.1016/j.bmcl.2011.05.082

Most ischemic stroke results from brain blood vessel blockage by platelet-mediated thrombus, and anti-platelet therapy has been demonstrated clinical benefits in the treatment of this disease. In the present work, novel nitric oxide (NO)-releasing derivatives of an anti-ischemic stroke drug 3-n-butylphthalide (NBP) were synthesized. Compounds 7a and 7c exhibited more potent anti-platelet activity than NBP and aspirin, and released a moderate amount of NO, which is beneficial in improving cardiovascular and cerebral circulation. These findings provide an alternative approach to the development of drugs more potent than NBP for the intervention of ischemic stroke.

Full text of DOI:10.1016/j.bmcl.2011.05.082

Bioorganic & Medicinal Chemistry Letters 21 (2011) 4210–4214
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Synthesis and evaluation of nitric oxide-releasing derivatives
of 3-n-butylphthalide as anti-platelet agents
Yang Li a, , Xuliang Wang a, , Rong Fu b, Wenying Yu a, Xiaoli Wang a, Yisheng Lai a,
Sixun Peng a, Yihua Zhang a,
⇑
a Center of Drug Discovery, China Pharmaceutical University, 24 Tongjiaxiang, Nanjing 210009, PR China
b Department of Physiology, China Pharmaceutical University, 24 Tongjiaxiang, Nanjing 210009, PR China
a r t i c l e i n f o
a b s t r a c t
Article history:
Most ischemic stroke results from brain blood vessel blockage by platelet-mediated thrombus, and anti-
platelet therapy has been demonstrated clinical benefits in the treatment of this disease. In the present
work, novel nitric oxide (NO)-releasing derivatives of an anti-ischemic stroke drug 3-n-butylphthalide
(NBP) were synthesized. Compounds 7a and 7c exhibited more potent anti-platelet activity than NBP
and aspirin, and released a moderate amount of NO, which is beneficial in improving cardiovascular
and cerebral circulation. These findings provide an alternative approach to the development of drugs
more potent than NBP for the intervention of ischemic stroke.
Received 29 March 2011
Revised 19 May 2011
Accepted 21 May 2011
Available online 27 May 2011
Keywords:
Ischemic stroke
Thrombus
Ó 2011 Elsevier Ltd. All rights reserved.
3-n-Butylphthalide
Nitric oxide-releasing derivatives
Anti-platelet activity
According to a report in 2004 from the World Health Organiza-
tion (WHO), 15 million people are suffering from stroke worldwide
annually. Of those, 5 million die, and another 5 million are perma-
nently disabled. Ischemic stroke is the most common cerebrovas-
cular disorder accounting for more than 80% of all strokes, a
majority of which are caused by platelet-mediated thrombosis.1
It is believed that thrombosis forms arterial blockage which leads
to inadequate blood supply to the brain tissue, and eventually re-
sults in ischemic stroke.2 It has been generally accepted that the
inhibition of platelets aggregation and adhesion plays a crucial role
in the treatment of ischemic stroke.3
The racemic 3-n-butylphthalide (NBP) received approval in Chi-
na for the treatment of ischemic stroke in 2002. A large number of
studies have demonstrated that NBP is able to inhibit platelet
aggregation, decrease the brain infarct volume, and enhance
microcirculation, thus benefiting patients with ischemic stroke.4
However, the clinical application is limited due to its moderate po-
tency. It has been found that the therapeutic effects of NBP could
be improved by co-administration with another anti-platelet
drug.5
It is well-known that nitric oxide (NO), a small endogenic gas
molecule, plays an important role in regulating physiological func-
tions, including the inhibition of platelet aggregation and throm-
bus formation in cerebrovascular and cardiovascular systems.6
During the past few years, our group and others have employed
a strategy by which an NO-donor moiety was connected to a ‘na-
tive’ molecule for the purpose of enhancing its therapeutic impact
and/or mitigating its adverse effects.7 Inspired by these successful
studies, we coupled compound 1 with NO-donors via various
linkers to generate NO-releasing derivatives of NBP (NO-NBP).
Compound 1 is a ring-opening derivative of NBP, which could un-
dergo a conversion to NBP very quickly in vivo.8 The linkers are cin-
namic acid derivatives such as ferulic acid and p-hydroxyl
cinnamic acid with both antioxidant and anti-platelet aggregation
activities.9 We expected that the ester bonds of NO-NBP could be
hydrolyzed in vivo by esterases to generate 1, which would subse-
quently undergo ring closure to produce NBP, and cinnamic acid
derivatives as well as NO (Scheme 1). These bioactive compounds
would possess synergistical inhibitory effects on platelet
aggregation.
Abbreviations: NO, nitric oxide; HOBT, 1-hydroxybenzotriazole; EDCI, 1-(3-
dimethyl aminopropyl)-3-ethylcarbodiimide hydrochloride; DMAP, 4-dimethyl-
aminopyridine; DCC, dicyclohexylcarbodiimide ; ADP, adenosine 50-diphosphate;
PRP, platelet rich plasma.
The synthetic routes of NO-NBP are outlined in Scheme 2. Feru-
lic acid 2a or p-hydroxyl cinnamic acid 2b was treated with alka-
nolamine nitrates 3a–d, respectively, in the presence of HOBT,
EDCI and DMAP at room temperature to give the corresponding ni-
trates 4a–h in 68–85% yields. Condensation of 4a–h with
⇑
Corresponding author. Tel./fax: +86 25 83271015.
E-mail address: zyhtgd@sohu.com (Y. Zhang).
These two authors contributed equally to this work.
0960-894X/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2011.05.082

Y. Li et al. / Bioorg. Med. Chem. Lett. 21 (2011) 4210–4214
4211
Scheme 1. Proposed metabolic route of NO-NBP.
Scheme 2. Synthetic routes of compounds 5a–h and 7a–g. Reagents and conditions: (i) 3, EDCI, HOBT, DMAP, CH2Cl2, rt, 5 h; (ii) 9a, DCC, DMAP, CH2Cl2, rt, 5 h; (iii) 9b, DCC,
DMAP, CH2Cl2, rt, 5 h; (iv) diethylamine, piperidine or morpholine, Et3N, DMF, rt, 8 h; (v) 1 M ethereal HCl, 0 °C, 1 h.
compound 9a in the presence of DCC and DMAP provided esters
5a–h in 78–83% yields.10 Similarly, treatment of 4a, 4e and 4f with
9b offered esters 6a–c in 75–85% yields. Compounds 7a–g were ob-
tained by reaction of 6a–c with diethylamine, piperidine and mor-
pholine, respectively, followed by treatment with anhydrous
ethereal HCl below 0 °C.11,12
ꢀ10 °C (Scheme 3). Compounds 9a and 9b were obtained starting
from reaction of 2-formylbenzoic acid 8 with Grignard reagent,
n-BuMgBr, followed by acidification to give NBP in 85% yields.
And subsequent ring-opening of NBP by saponification formed acid
1 in 92% yield. Finally, acylation of hydroxyl at the side chain of 1
with acetyl chloride or chloroacetyl chloride provided 9a and 9b in
yields of 77% and 83%, respectively. In addition, compound 10 was
obtained by condensation of 9b with diethylamine in 62% yield
The alkanolamine nitrates 3a–d were prepared by treatment of
corresponding alkanolamine with fuming HNO3 and Ac2O at

4212
Y. Li et al. / Bioorg. Med. Chem. Lett. 21 (2011) 4210–4214
Scheme 3. Synthetic routes of compounds 3a–d, 9a, 9b and 10. Reagents and conditions: (i) (1) n-BuMgBr, Et2O, –5–0 °C, 5 h; (2) 1 M HCl, RT, 0.5 h; (ii) (1) NaOH, CH3OH–
H2O, reflux, 0.5 h; (2) 1 M HCl, ꢀ10 to 0 °C; (iii) ClCH2COCl, Et3N, DMAP, CH2Cl2, ꢀ10 °C; (iv) CH3COCl, Et3N, DMAP, CH2Cl2, ꢀ10 °C. (v) diethylamine, Et3N, CH2Cl2, rt, 12 h; (vi)
HNO3, Ac2O, CH2Cl2, ꢀ10 °C; (vii) HNO3, Ac2O, Et2O, ꢀ10 °C.
(Scheme 3). All target products were purified by column chroma-
tography and their structures were identified by IR, MS, 1H NMR,
and HR-MS.
platelet aggregation with IC50 values of 54.44
respectively.14
lM and 39.40 lM,
Since the structure of compound 7a is composed of two large
moieties, NBP precursor 10 and NO donor 4a as shown in Chart 1,
we further investigated their individual contribution to the overall
activity against ADP-induced platelet aggregation in vitro. We
found that both 10 and 4a displayed significant anti-platelet activ-
ity (67.2% and 51.3%, respectively), although each moiety was less
active than the parent compound 7a (81.2%), indicating that these
two moieties in 7a may synergistically inhibit the ADP-induced
platelet aggregation. In addition, the anti-platelet effect of 7a
The inhibitory effects of the target compounds on adenosine 50-
diphosphate (ADP)-induced platelet aggregation in rabbit platelet
rich plasma (PRP) were evaluated by using Born’s turbidimetric
method.13 As shown in Figure 1, several compounds (7a, 7c, 7f,
and 7g) at 1.0 mM inhibited ADP-induced platelet aggregation by
83.0%, 83.0%, 82.4% and 82.1%, respectively, significantly stronger
than control drugs, NBP (67.8%) and aspirin (ASP) (68.5%). And
the most potent compounds 7a and 7c inhibited ADP-induced
was dramatically attenuated by treatment with 20 lM of hemoglo-
bin (He), an endogenous NO scavenger,15 resulting in reduction in
inhibition from 81.2% to 54.4% (Fig. 2). These results strongly sup-
port the important role of NO-releasing moiety for anti-platelet ef-
fects of NO-NBP.
In order to further explore the relationship between NO re-
leased by NO-NBP and their anti-platelet aggregation, the levels
Figure 1. Inhibition of ADP-induced rabbit platelet aggregation by target com-
pounds in vitro. Rabbit platelet suspensions were pre-incubated with tested
compounds (1.0 mM) at 37 °C for 5 min and exposed to 10 lM of ADP, followed by
continually monitoring. Rabbit platelet suspensions that had been treated with
vehicle and exposed to ADP were used as positive controls. Data are expressed as
mean S.D. of each group (n = 4) from several independent experiments. ⁄P < 0.05
versus NBP, #P < 0.05 versus ASP, determined by Student’s t test.
Chart 1. Structure of compound 7a.

Y. Li et al. / Bioorg. Med. Chem. Lett. 21 (2011) 4210–4214
4213
anti-platelet activity. For example, compounds containing amino-
ethyl are generally more active than those containing aminopropyl
(5a vs 5b, 7a vs 7g). It is likely that those various substituents and
linkers may have different abilities to modulate the structure, sta-
bility, metabolism and penetrability, which affect the NO produc-
tion and NBP releasing, leading to varied bioactivities of NO-NBP.
Indeed, Compound 7a containing both diethylamino and amino-
ethyl moieties produced higher level of NO and displayed strongest
inhibitory activity among the compounds tested. These prelimin-
ary SAR results provide a guideline for further design and develop-
ment of NO-releasing derivatives of NBP.
Taken together, a novel class of NO-NBP hybrids were synthe-
sized based on our previous studies, and all the target compounds
exhibited inhibitory effects on platelet aggregation to various ex-
tents. Apparently, the synergic effects of NBP and NO production
contribute to their potent activity against platelet aggregation.
Compounds 7a and 7c, which had a great potency superior to
ASP and NBP, may be promising candidates for further intensive
study. Furthermore, several NO-NBP produced moderate levels of
NO, which are correlated with their anti-platelet aggregation activ-
ities. Interestingly, our previous pharmacokinetic study showed
that the compound structurally similar to 7a and 7c could produce
the parent molecule NBP under physiological conditions, which is
in accordance with our working assumption.18 Notably, the two
moieties of 7a, NBP precursor 10 and NO donor 4a, had moderate
anti-platelet aggregation activity, and treatment with NO quencher
mitigated the inhibitory effect of 7a on the ADP-induced platelet
aggregation. These results indicate that these two moieties may
perform synergistical inhibition on platelet aggregation. As a re-
sult, our novel findings may provide a novel framework for the de-
sign of new NO-releasing NBP hybrids for the intervention of
ischemic stroke.
Figure 2. Inhibition of ADP-induced rabbit platelet aggregation by selected
compounds in vitro. Rabbit platelet suspensions were pre-incubated with tested
compounds (1.0 mM) at 37 °C for 5 min followed by addition of ADP (10
lM).
Hemoglobin (20 M) was added and incubated with the drug-platelet suspension in
l
the indicated group. Data are expressed as mean SD from several independent
experiments. ⁄P < 0.05 versus NBP, determined by Student’s t test.
of NO produced by 5c, 5e, 7a, 7c and 7e were evaluated by Griess
assay.16 It is shown in Figure 3 that the active compounds 7a, 7c
and 7e released moderate amount of NO (0.057, 0.060, and
0.043
5c and 5e under the same conditions released lower levels of NO
(0.027 and 0.022 g/ml, respectively). These date suggest that
lg/ml, respectively). In contrast, the less active compounds
l
the NO-releasing ability of NO-NBP may be correlated to their
inhibitory effects on platelet aggregation (R2 = 0.98, P < 0.01, deter-
mined by logistic regression analysis).
Analysis of structure–activity relationships (SAR) revealed that
the target compounds with different acylation of hydroxyl in the
side chain of NO-NBP exhibited varied inhibitory activities on
platelet aggregation. Acylation by diethylaminoacetyl or piperidi-
noacetyl functional group generally led to higher activity than
acetylation (7a vs 5a, 7f vs 5e, 7g vs 5f), probably owing to en-
hanced aqueous solubility and improvement of cell permeability,
since our previous studies showed that some NO-NBP compounds
structurally similar to 7a had a greatly improved aqueous solubil-
ity in comparison with NBP, and displayed significant capacity to
penetrate across blood–brain barrier to retard cerebral injury after
ischemia–reperfusion.17 On the other hand, the carbon chain
length of aminoalkyl connecting to carboxyl of ferulic acid or p-hy-
droxyl cinnamic acid moiety of NO-NBP seems correlated to their
Acknowledgments
This study was supported by a grant from the Major National
Science and Technology Program of China for Innovative Drug dur-
ing the Eleventh Five-Year Plan Period (No. 2009ZX09103-095).
References and notes
1. Lloyd, J. D.; Adams, R.; Carnethon, M. Circulation 2009, 119, e21.
2. Stoll, G.; Kleinschnitz, C.; Nieswandt, B. Blood 2008, 112, 3555.
3. Aleviadou, B. R.; Moake, J. L. Blood 1993, 81, 1263.
4. (a) Liu, C.; Liao, S.; Zeng, J.; Lin, J.; Li, C.; Xie, L.; Shi, X.; Huang, R. J. Neurosci.
2007, 260, 106; (b) Zhu, X.; Li, X.; Liu, J. Eur. J. Pharmacol. 2004, 500, 221; (c)
Yan, C.; Feng, Y.; Zhang, J. Acta Pharmacol. Sin. 1998, 19, 117; (d) Peng, Y.; Zeng,
X.; Feng, Y.; Wang, X. J. Cardiovasc. Pharmacol. 2004, 43, 876.
0.07
5. Xu, H. L.; Feng, Y. P. Acta Pharm. Sin. 2001, 36, 329.
6. Koh, K. K. Int. J. Cardiol. 2003, 87, 1.
5c
0.06
0.05
0.04
0.03
0.02
0.01
0
7e
5e
7a
7c
7. (a) Fang, L.; Appenroth, D.; Decker, M.; Kiehntopf, M.; Zhang, Y.; Lehmann, J. J.
Med. Chem. 2008, 51, 713; (b) Huang, Z.; Zhang, Y.; Zhao, L. Org. Biomol. Chem.
2010, 8, 632; (c) Ling, Y.; Ye, X.; Zhang, Y. Bioorg. Med. Chem. 2010, 18, 3448.
8. Zhang, Y.; Wang, L.; Zhang, L.; Wang, X. Drug Dev. Res. 2004, 63, 174.
9. (a) Fang, J.; Zhang, Y.; Chang, X.; Zhao, B.; Jiang, D.; Saito, M.; Li, Z. J. Agric. Food.
Chem. 2009, 57, 8683; (b) Maurya, D. K.; Devasagayam, T. P. Food Chem.Toxicol.
2010, 48, 3369.
10. General procedure for the synthesis of the target compounds 5a–h: DMAP
(0.1 mmol) and appropriate nitrates 4a–h (0.31 mmol) were added to
a
mixture solution of acetyl ester 9a (0.26 mmol) and DCC (0.33 mmol) in dry
CH2Cl2 (25 mL). The reaction mixture was stirred at room temperature for 24 h.
Filtration and removal of the solvent in vacuo afforded the crude product,
which was subsequently purified by column chromatography using (PE/
EtOAc = 2:1) to give pure 5a–h in 78–83% yields.
0
50
100
150
200
250
300
350
Time (min)
11. General procedure for the synthesis of the target compounds 7a–g: A solution of
chloroacetyl ester 9b (1.0 mmol), DCC (1.25 mmol), DMAP (0.125 mmol) and
appropriate nitrates 4a, 4e and 4f (1.1 mmol) in dry CH2Cl2 (25 mL) was stirred
at room temperature for 24 h. After filtration, the filtrate was evaporated in
vacuo and the crude product was purified by column chromatography (PE/
Figure 3. NO-release assay of selected compounds in vitro. NOꢀ2 concentrations
which represent the quantity of NO were determined by Griess assay. Griess
reagent can combine with NOꢀ2 and form chromospheres after 10 min at 30 °C. The
absorbance then was measured at 540 nm. The levels of NO produced by individual
compounds were calculated, according to the standards of different concentrations
of nitrate. Data are expressed as means of individual compounds tested at each time
point and intra-group variations were less than 10%.
EtOAc = 3:1) to afford esters 6a–c.
A solution of appropriate amines
(0.45 mmol), Et3N (0.45 mmol) and 6a (0.15 mmol) in DMF (15 mL) was
stirred at room temperature for 8 h. The solvent was reconstituted in EtOAc
and water, and the organic phase was washed with water and brine dried over

4214
Y. Li et al. / Bioorg. Med. Chem. Lett. 21 (2011) 4210–4214
anhydrous Na2SO4, and concentrated under reduced pressure. The crude
centrifuging at 500 rpm for 10 min. The supernatants were collected and used
as platelet rich plasma (PRP). Additional samples were centrifuged at 3000 rpm
for 10 min and the supernatants were collected as platelet poor plasma (PPP).
The effect of individual compounds on the ADP-induced platelet aggregation
was measured by the Born’s turbidimetric method using a Platelet-
Aggregometer (LG-PABER-I Platelet-Aggregometer, Beijing). Briefly, PRP
(240 l) was pre-treated in duplicate with vehicle, different concentrations of
individual compounds or the reference drugs for 5 min and exposed to 10 lM
of ADP incubated at 37 °C for 5 min. The formation of platelet aggregation was
monitored longitudinally by optical density. Platelet aggregation was induced
by ADP (final concentration 10 lM). Compounds under study or vehicle alone
were added to the PRP samples 5 min before addition of the aggregating agent.
The antiplatelet aggregation activity of tested compound individual
compounds was evaluated as percent inhibition of platelet aggregation
compared to positive controls that had been pre-treated with vehicle alone
and exposed with the inducer samples. For most active compounds, The IC50
values of most compounds were determined by nonlinear regression analysis
and the percent inhibition at the maximal concentration tested (1.0 mM) was
calculated.
product was dissolved in EtOAc (5 mL) and 1 M ethereal HCl (0.5 mL), and left
stirring at 0 °C for 1 h to yield pure hydrochlorides 7a–c. The target compound
7d–g was obtained under the same procedure.
12. Analytical data for 7a: mp 103–105 °C. ESI–MS: m/z 586 [M+H]+. IR (KBr): 757,
1035, 1630, 1743, 2957, 3067 cmꢀ1
.
1H NMR (CDCl3, 500 Hz, d): 0.78 (t, 3H,
CH3, J = 6.6 Hz), 1.02 (m, 7H, 2ꢁ CH2 and CH3), 1.22 (t, 3H, CH3, J = 7.2 Hz), 1.81
(m, 2H, CH2), 3.22 (m, 4H, 2ꢁ NCH2), 3.33 (m, 2H, NCH2), 3.63 (m, 2H, COCH2N),
3.78 (s, 3H, OCH3), 4.51 (t, 2H, CH2ONO2, J = 7.0 Hz), 6.29 (d, 1H, CH@,
J = 16.0 Hz), 6.57 (m, 1H, CH), 6.75 (m, 1H, CONH), 7.03 (m, 3H, ArH), 7.35–7.55
(m, 3H, ArH), 7.60 (d, 1H, CH@, J = 15.9 Hz), 8.07 (d, 1H, ArH, J = 7.8 Hz), 10.58
(br s, 1H, NH+). 13C NMR (CDCl3, 500 Hz, d): 165.8, 165.4, 164.3, 151.0, 144.2,
140.1, 138.3, 134.3, 133.5, 130.5, 128.1, 126.6, 126.4, 123.2, 122.4, 120.2, 111.7,
74.1, 72.4, 64.8, 55.9, 51.2, 48.2, 43.3, 40.9, 35.7, 27.2, 21.6, 13.6, 8.7. HR-MS for
C
30H40N3O9 ([M+H]+) calcd 586.2765, found: 586.2768. 7c: mp 113–114 °C.
ESI–MS: m/z 598 [M+H]+. IR (KBr): 755, 1037, 1623, 1739, 2970, 3069 cmꢀ1 1H
.
NMR (CDCl3, 500 Hz, d): 0.85 (t, 3H, CH3, J = 6.3 Hz), 1.15 (m, 4H, 2ꢁ CH2),
1.86ꢂ1.92 (m, 8H, 4ꢁ CH2), 3.5 (m, 4H, 2ꢁ NCH2), 3.68 (m, 2H, NCH2), 3.81 (m,
2H, COCH2N), 3.85 (s, 3H, OCH3), 4.60(t, 2H, CH2ONO2, J = 6.8 Hz), 6.37 (d, 1H,
CH = , J = 16.0 Hz), 6.65 (m, 1H, CH), 6.78 (m, 1H, CONH), 7.05–7.20 (m, 3H,
ArH), 7.39 (m, 1H, ArH), 7.56 (d, 1H, CH = , J = 15.9 Hz), 7.62–7.67 (m, 2H, ArH),
8.15 (d, 1H, ArH, J = 7.7 Hz), 10.20 (br s, 1H, NH+). 13C NMR (CDCl3, 500 Hz, d):
165.3, 165.1, 164.2, 150.9, 142.1, 139.9, 138.3, 134.2, 133.5, 130.5, 128.0, 126.6,
126.3, 123.1, 122.3, 120.2, 111.7, 73.9, 64.8, 55.9, 55.5, 52.8, 48.2, 43.4, 40.8,
15. Gong, P.; Cederbaum, A. I.; Nieto, N. Mol. Pharmacol. 2004, 65, 130.
16. Nitrate/nitrite measurement in vitro: Briefly, 0.1 mM of each compound in
phosphate buffer solution (PBS) containing 2% dimethyl sulfoxide and 5.0 mM
L
-cysteine at pH 7.4 was incubated at 37 °C for 15–300 min and were sampled
every 15 min for 120 min and then every 30 min for the remaining time. The
collected samples (2 mL) were mixed with 0.5 ml of Griess reagent and
incubated at 37 °C for 10 min, followed by measuring at 540 nm. The different
concentrations of nitrite were used as standards to calculate the
concentrations of NO formed by individual compounds.
35.7, 27.2, 21.9, 21.6, 21.0, 13.6. HR-MS for
598.2765, found: 598.2762.
C
31H40N3O9 ([M+H]+) calcd
13. (a) Born, G. V.; Cross, M. J. J. Physiol. 1963, 168, 178; (b) Jones, M.; Inkielewicz,
I.; Medina, C.; Santos-Martinez, M. J.; Radomski, A.; Radomski, M. W.; Lally, M.
N.; Moriarty, L. M.; Gaynor, J.; Carolan, C. G.; Khan, D.; O’Byrne, P.; Harmon, S.;
Holland, V.; Clancy, J. M.; Gilmer, J. F. J. Med. Chem. 2009, 52, 6588.
14. Antiplatelet aggregation assays: Blood samples were withdrawn from rabbit
carotid artery and mixed with 3.8% trisodium citrate (9:1 v/v), followed by
17. Zhuang, P.; Ji, H.; Zhang, Y. Clin. Exp. Pharmacol. Physiol. 2010, 37, e121.
18. Li, N.; Wang, X.; Li, T.; Ji, H.; Zhang, Y.; Qiu, Z.; Zhao, D.; Chen, X. Xenobiotica
2011, accepted, DOI: 10.3109/00498254.2011.580385.