全 文 ： 164 Chin J Nat Med Mar. 2013 Vol. 11 No. 2 2013年 3月 第 11卷 第 2期

Chinese Journal of Natural Medicines 2013, 11(2): 01640170
doi: 10.3724/SP.J.1009.2013.00164
Chinese
Journal of
Natural
Medicines


Development of a hydrophilic interaction
chromatography-UPLC assay to determine trigonelline in rat
plasma and its application in a pharmacokinetic study
CHENG Zai-Xing1,4, WU Jin-Jun3, LIU Zhong-Qiu2*, LIN Na1,4*
1 College of Pharmacy, Fujian University of Traditional Chinese Medicine, Fuzhou 350108, China;
2 Department of Pharmaceutics, School of Pharmaceutical Sciences, Southern Medical University, Guangzhou 510515, China;
3 School of Chinese Materia Medica, Guangzhou University of Chinese Medicine, Guangzhou 510006, China;
4 Institute of Chinese Meteria Medica, China Academy of Chinese Medical Sciences, Beijing 100700, China
Available online 20 Mar. 2013
[ABSTRACT] AIM: Trigonelline (Tr) is the second most abundant alkaloid in coffee beans. This study developed an assay combining
hydrophilic interaction chromatography with ultra performance liquid chromatography (HILIC-UPLC) for the quantification of Tr in
rat plasma to determine its pharmacokinetic behavior. METHODS: After the administration of Tr by gavage as well as intravenous
injection and that of methanol extract of coffee beans (MECB) orally, blood samples from the experimental rats were analyzed using
the HILIC-UPLC assay. Pharmacokinetic parameters were determined using the standard non-compartmental method and calculated
using Practical Pharmacokinetic Program Version 87/97. RESULTS: The HILIC-UPLC assay was validated with the linear range of
0.12–100 g·mL1 and a lower limit of quantitation of 0.12 g·mL1. Its accuracy, precision, recovery, and stability were within
acceptable limits. The AUC(0∞) (where AUC is the area under the plasma concentration–time curve) values were determined to be
(4 066.83  1 244.41) and (3 544.29  908.80) min·g·mL1 after Tr was orally and intravenously administered, respectively. It was
(4 566.75  1 435.64) min·g·mL1 after MECB was orally administered. The absolute bioavailability of Tr alone reached 57.37%,
whereas that of Tr in MECB was 64.42%. The relative bioavailability of the alkaloid was 112.29%. CONCLUSIONS: The
HILIC-UPLC assay for Tr determination is simple and accurate, and also exhibits good reproducibility. The bioavailability of
stand-alone Tr and that of Tr in MECB were both good. Tr alone and that in MECB orally administered did not exhibit any significant
difference.
[KEY WORDS] HILIC-UPLC assay; Trigonelline; Coffee beans; Bioavailability
[CLC Number] R917 [Document code] A [Article ID] 1672-3651(2013)02-0164-07

1 Introduction
Coffee is one of the three main drinks in the world, and
is popular in Europe, the United States, and China, among
other countries. Much research about coffee and its effects on
health has been carried out, with caffeine being widely

[Received on] 14-May-2012
[Research funding] This project was supported by the National
Basic Research Program of China (973 Program) (Nos. 2009CB5228008,
and 2011CB505300/2011CB5305) and the National Natural Science
Foundation of China (No. U0832002).
[*Corresponding author] LIU Zhong-Qiu: Prof., Tel: 86-20-61648715,
Fax: 86-20-61648596, Email: liuzq@smu.edu.cn; LIN Na: Prof., Tel:
86-10-66401441,E-mail:linna888@163.com
These authors have no conflict of interest to declare.
considered as the main ingredient. Trigonelline (Tr)
(1-methyl-3-carboxy-pyridinium ion), also known as
N-methylnicotinic acid, is the second most abundant alkaloid
in coffee beans. Almost all green coffee beans have high
levels of Tr [1]. This alkaloid is also the major active and
nutrient ingredient of coffee beans, and the main source of
important volatile coffee flavor components in roasting [2].
Tr, a vitamin B3 precursor [3], is an alkaloid belonging to
a group of pyridine betaines possessing a quaternary amino
group. Several health-promoting properties of Tr, such as
hypoglycemic, hypocholesterolemic, antitumor, antimigraine,
memory-improving, and antiseptic activities, have been
reported [45]. However, Tr has also been reported to be able
to stimulate the growth of estrogen-dependent MCF-7 breast
cancer cells in vitro [6]. Thus, a Tr-rich diet might play a role
CHENG Zai-Xing, et al. /Chinese Journal of Natural Medicines 2013, 11(2): 164170
2013年 3月 第 11卷 第 2期 Chin J Nat Med Mar. 2013 Vol. 11 No. 2 165

in the development of cancer [2]. In addition to coffee, Tr is
distributed widely in the plant and animal kingdoms [7], for
instance, in tea, pumpkin, beans, Sardinia melanosticta, and
Rhopilema esculenta, as well as in traditional Chinese
medicine agents, such as Trigonella foenum-graecum L.,
Pinellia ternata (Thunb.) Ten. ex Breitenb., and Radix
mirabilis. Nowadays, Tr is being principally used to bring
down a fever, relieve a cough, lower cholesterol levels, and
regulate blood sugar levels. It is hence an ingredient closely
linked to the daily diet and with a significant impact on
human health.
Tr has been extensively studied in many aspects. An
HPLC method was reported for determining Tr in rat plasma
with an aminopropyl column [8]. However, Tr is a strong
polar hydrophilic compound with anions and cations. It could
not be fully separated by an aminopropyl column [9]. Tr has
little retention on a C18 column with reversed-phase liquid
chromatography [10]. It takes 3.5 h to equilibrate the column
when reversed-phase ion-pair chromatography is used to
separate Tr [11]. By contrast, hydrophilic interaction
chromatography (HILIC), which contains a strong polar
stationary phase, as well as a mobile phase with a high
concentration of organic solvent and a low concentration of
aqueous eluent, is a novel analytical method adapted for
strong polar components [11]. It is especially suitable for
situations in which compound retention is poor in
reversed-phase liquid chromatography. Tr has been found to
have good retention on a HILIC column, and a low limit of
quantitation on ultra performance liquid chromatography
(UPLC) [11-12].
The objective of this study was thus to develop and
validate a HILIC-UPLC assay for determining the presence
of Tr in rat plasma that could be used in routine
pharmacokinetic studies investigating the bioavailability of
Tr in the methanol extract of coffee beans (MECB) using rats.
This method allows for further research on the health effects
of Tr in coffee beans.
2 Experimental
2.1 Materials
Tr (purity,  98%) was purchased from Chengdu Mansite
Biotechnology Co., Ltd. (Sichuan Province, China). Green
coffee beans, identified as the semen of Coffea arabica L.
(Yunnan Province, China), were purchased from Fuzhou
Zefan Restaurant Management Co. Ltd. HPLC-grade
methanol and acetonitrile were purchased from Merck & Co.
Inc. (USA). HPLC-grade ammonium acetate was purchased
from J.T. Baker (USA). HPLC-grade water processed by a
Milli-Q system (Millipore, USA) was used. All other
chemicals were of analytical grade. Blank rat plasma (drug
free) was prepared in this laboratory.
2.2 Animals
Male, pathogen-free Sprague-Dawley rats (250–300 g)
were provided by the Guangzhou University of Chinese
Medicine Experimental Animal Center (Guangzhou, China).
Animal studies were carried out in accordance with the
Guidelines for Animal Experimentation of Southern Medical
University (Guangzhou, China), and the procedure was
approved by the Animal Ethics Committee of the same
institution.
2.3 Preparation of the extract
The coffee beans were extracted with 20-fold mass of
50% methanol using an ultrasonic apparatus for 60 min and
then filtered. After the methanol was removed under reduced
pressure, the residue was dried in a vacuum oven.
The methanol extract of coffee beans (MECB) was
redissolved in 50% methanol and centrifuged for 30 min at
13 000 r·min1. The supernatant (10 L) was subjected to
UPLC analysis. The analytical method was the same as
described in the “2.6. analytical procedures” section. The
results indicated that the content of Tr was 29.13 mg·g1 in
MECB.
2.4 Collection of plasma samples
Blood samples (0.3-0.5 mL) were taken from the
retro-orbit sinus before experimentation, and at 2, 10, 30, 60,
90, 120, 180, 240, 360, 480, 720, and 1 440 min after oral or
intravenous administration of Tr and oral administration of
MECB. Plasma specimens were separated by centrifugation
at 8 000 r·min1 for 8 min, stored at 4 C, extracted within 6 h,
and incubated at 20C.
2.5 Preparation of samples
Rat plasma (100 L) was mixed with 10 L of methanol
or the working standard solution by vortexing vigorously for
30 sec at 3 000 r·min1 at ambient temperature. Next, 3  100
L of acetonitrile was added to the mixture during vortex
mixing for 90 s to precipitate protein. The samples were
centrifuged at 13 000 r·min1 for 30 min. Finally, 360 L of
the supernatant was transferred to a clean tube, the air
evacuated, and the sample stored at 20C until analysis. The
samples were redissolved with 180 L of 70% acetonitrile
aqueous solution, vortexed for 3 min at 3 000 r·min1, and
centrifuged for 30 min at 13 000 r·min1. The supernatant (10
L) was subjected to UPLC analysis.
2.6 Analytical procedures
UPLC was performed using a Waters Acquity UPLC
System (Waters, USA) and PDAeλ Detector. Separations
were achieved with a Halo HILIC column (2.1 mm  100
mm, 2.7 m) at 30 C. The mobile phase consisted of
acetonitrile/2.5 mol·L1 aqueous ammonium acetate (70 : 30,
V/V) with a flow rate of 0.4 mL·min1 in a run time of 2.2
min. The UV detection wavelength was set at 265 nm.
2.7 Preparation of calibration standard and quality control
(QC) samples
A stock solution of Tr (1 mg·mL1) was prepared in
methanol and stored at 20 C. The stock solution was
CHENG Zai-Xing, et al. /Chinese Journal of Natural Medicines 2013, 11(2): 16470
166 Chin J Nat Med Mar. 2013 Vol. 11 No. 2 2013年 3月 第 11卷 第 2期

diluted quantitatively with methanol to give working
standards at concentrations of 1.17, 2.34, 4.69, 9.38, 37.5,
150, 600, and 1 000 g·mL1. Calibration standards of Tr
were prepared by dissolving the working standards with
blank rat plasma, yielding the final concentrations of 0.12,
0.23, 0.47, 0.94, 3.75, 15, 60, and 100 g·mL1, respectively.
QC samples were prepared at low (0.23 g·mL1), medium
(3.75 g·mL1), and high (60 g·mL1) concentrations in the
same way, as were the plasma samples of calibration, to
evaluate the accuracy, precision, recovery, and stability of the
assay.
2.8 Method validation
2.8.1 Specificity
Specificity was assessed by analyzing six samples of
blank rat plasma with and without spiking with Tr.
2.8.2 Linearity and lower limit of quantitation (LLOQ)
Calibration curves were constructed from working
standard solutions of Tr at concentrations ranging from 0.12
to 100 g·mL1 by plotting the peak area (y) of Tr versus its
concentration (x) with 1/x2 weighted regression. The assay
was evaluated with an eight-point calibration plot at the
above-described concentration range.
The LLOQ was defined as the lowest concentration in
the calibration curve that could be determined with accuracy
and precision of no more than 20%.
2.8.3 Accuracy and precision
Accuracy (determined as the percentage of difference
between the mean detected concentrations and the nominal
concentrations) as well as intra- and inter-day precision
(expressed as relative standard deviation, RSD) were
assessed by assay of six replicate QC samples on 3 different
days.
2.8.4 Extraction recovery
The extraction recovery of Tr was determined by
assaying two batches of samples: plasma extracts spiked with
Tr after extraction (batch 1) and plasma spiked with Tr before
extraction (batch 2). The Tr of each batch was prepared at
levels of 0.23, 3.75, and 60 g·mL1 and in sets of six
replicates. The extraction recovery of Tr was calculated as
follows: Extraction Recovery = Batch 2/Batch 1  100% [13].
2.8.5 Stability
The stability of Tr in rat plasma was evaluated by
analyzing replicates (n = 5) of 0.23, 3.75, and 60 g·mL1,
which were exposed to different time and temperature
conditions. Short-term stability was determined after the
exposure of the spiked samples at approximately 25 C for 6
h, and the stability of the post-extracted samples was
evaluated at 20 C for 3 d. Long-term stability was assessed
after storage of the samples at 20 C for 2 weeks, and the
stability of the redissolved samples was evaluated at
approximately 4 C for 12 h. The samples at concentrations
following storage were compared with freshly prepared
samples at the same concentrations [14].
2.9 Pharmacokinetic study
Twenty Sprague-Dawley rats were fasted with water
access for 12 h prior to the initiation of the study and
randomly divided into four groups: One group was given
normal saline to quantify the content of endogenous Tr,
another group was intravenously administered with Tr at the
dose of 5 mg·kg1, yet another group was orally administered
with Tr at the dose of 10 mg·kg1, and the remaining group
was orally administered with MECB at the dose of 10 mg·kg1.
Plasma samples were prepared as the above-described method.
2.10 Statistical analysis
Data in the present study were presented as mean ± SD.
Student’s t-test was used to evaluate statistical differences. P
< 0.05 were considered statistically significant.
Pharmacokinetic parameters were estimated by a
non-compartmental method from the plasma concentration–
time data using Practical Pharmacokinetic Program Version
87/97. The maximum plasma concentration (cmax) and time to
peak concentration (Tmax) were obtained from the observed
data. The absolute bioavailability (AF) of Tr after an oral
administration was calculated as follows: [AUC(oral administration)
 Dose(intravenous injection)]/[AUC(intravenous injection)  Dose(oral administration)]
 100%, where AUC is the area under the plasma
concentration–time curve. The relative bioavailability of Tr
after an oral administration of MECB was calculated as
AUC(MECB)/AUC(Tr)  100%.
3 Results and Discussion
3.1 Stability of stock solution
As Tr is freely soluble in methanol, it was dissolved in
methanol to prepare the stock solution. The mean values of 1
mg·mL1 stock solution of Tr at ambient temperature for 6 h
and at 20 C for 30 d were (0.99  0.01) mg·mL1 (RSD =
1.32%) and (0.97  0.01) mg·mL1 (RSD = 1.13%), respec-
tively. The results showed that Tr was stable in methanol
under the stated conditions.
3.2 Solubility of Tr in the mobile phase
Tr is nearly insoluble in acetonitrile, but it was found to
be soluble in the mobile phase with the 70% acetonitrile
solution used for redissolution. Hence, the solubility of Tr in
70% acetonitrile aqueous solution was investigated. The
results showed that its solubility was in excess of 2 mg·mL1.
3.3 Method validation
3.3.1 Specificity
Typical chromatograms of the sample solution, blank
plasma, blank plasma spiked with Tr, and rat plasma samples
after administration are presented in Fig. 1. Under the elution
conditions, Tr was well separated at the retention time of 1.83
min. The peak of Tr was exactly symmetrical.
3.3.2 Linearity and LLOQ
The linear regression of the calibration curve for the
peak area of Tr (y) versus the concentration of Tr (x) was
plotted for plasma samples and fitted over the concentration
CHENG Zai-Xing, et al. /Chinese Journal of Natural Medicines 2013, 11(2): 164170
2013年 3月 第 11卷 第 2期 Chin J Nat Med Mar. 2013 Vol. 11 No. 2 167

range of 0.12–100 g·mL1. For the determination of Tr in rat
plasma, the regression equation of y = 10 928 x  2 428.3 and
correlation coefficient (r) of 0.999 9 were obtained. The
LLOQ of Tr was 0.12 g·mL1.
3.3.3 Precision and accuracy
The results of intra- and inter-day precision and accuracy
are shown in Table 1. The intra- and inter-day precision values
(RSDs) were within the acceptable range of 15%, whereas
the assay accuracies ranged from 95.84% to 103.48%. The
method proved to be highly accurate and precise.
3.3.4 Extraction recovery
The extraction recovery rates for Tr were (93.59  7.88)%,



Fig. 1 Typical chromatograms of the (A) sample solution, (B) blank plasma, (C) blank plasma spiked with Tr, (D)
plasma after Tr intravenous administration, and (E) plasma after Tr oral administration.
(B) is the chromatographic peak of endogenous Tr; (C) is the chromatographic peak of 3.75 g·mL1 Tr; (D) is the
chromatographic peak of Tr at the first blood drawing point after its intravenous administration at 5 mg·kg1; (E) is the
chromatographic peak of Tr at the first blood drawing point after its oral administration at 10 mg·kg1

Table 1 Precision and accuracy of the HILIC-UPLC assay for the quantification of Tr in rat plasma (n = 6)
Sample concen-Tration/(g·mL1) Average measured concentration/(g·mL1) Intra-day RSD/% Inter-day RSD/% Accuracy/%
60 62.09 2.49 4.48 103.48
3.75 3.59 3.15 4.32 95.84
0.23 0.26 13.72 6.05 95.98

CHENG Zai-Xing, et al. /Chinese Journal of Natural Medicines 2013, 11(2): 16470
168 Chin J Nat Med Mar. 2013 Vol. 11 No. 2 2013年 3月 第 11卷 第 2期

(88.23  3.42)%, and (80.09  3.32)% at the concentrations
of 0.23, 3.75, and 60 g·mL1, respectively.
3.3.5 Stability
Stability data for Tr in plasma are summarized in Table 2.
Tr was stable after rat plasma was placed at ambient tem-
perature (about 25 C) for 6 h and stored at 20C for 2
weeks, after the post-extracted sample was stored at
approximately 20 C for 3 d, and after the redissolved
sample was stored at approximately 4C for 12 h. The results
showed a reliable stability behavior of Tr under the
experimental conditions.

Table 2 Stability of Tr in rat plasma under different conditions as determined by HILIC-UPLC assay (n = 5)
Condition Sample concentration
c/(g·mL1)
Average measured concentration
c/(g·mL1)
Accuracy/% Precision
RSD/%
60 55.26 92.10 9.61
3.75 3.36 89.46 10.90

Short-term stability
(about 25 C, 6 h)
0.23 0.22 94.07 10.21
60 67.80 112.99 3.17
3.75 3.89 103.78 5.25

Long-term stability
(20 C, 2 weeks)
0.23 0.22 95.50 7.19
60 65.09 108.48 14.73
3.75 4.24 113.09 6.04

Stability of the post-extracted
samples (20 C, 3 d)
0.23 0.23 99.83 8.99
60 64.24 107.06 10.48
3.75 4.21 112.27 12.90 Stability of the redissolved samples (about 4 C, 12 h)
0.23 0.24 103.10 11.82

In brief, the specificity, accuracy, precision, recovery,
and stability tests all met the requirements for the quantitative
biological samples. The HILIC-UPLC assay fitted well with
the data obtained for Tr in rat plasma.
3.4 Pharmacokinetic study
As illustrated in Fig. 1B, endogenous Tr was present in
blank plasma, the content of which varied with the time
points of blood drawing in five rats. Rat plasma was replaced
with normal saline to construct a calibration curve. A mixture
of plasma from each rat (100 L) was prepared following the
above-described method to enhance the limit of detection [15].
The results showed that the content of endogenous Tr ranged
from 0.02 to 0.26 g·mL1. Hence, in the pharmacokinetic
studies, all data were subtracted from the mean content of
endogenous Tr at corresponding time points of blood
drawing.
The assay was successfully applied in the determination
of Tr in this preclinical pharmacokinetic study. The mean
plasma concentration–time profile is depicted in Fig. 2,
which showed that the concentrations of Tr in rat plasma
were quantifiable at least 24 h after gavage and intravenous
administration. The main pharmacokinetic parameters are
reported in Table 3. After gavage and intravenous admini-
stration of Tr, the AUC(0-∞) values were (4 066.83  1 244.41)
and (3 544.29  908.80) min·g·mL1, the t1/2 (biological
half-life) values were (215.86  15.91) and (201.49  37.33)
min, and the CL(s) (clearance) values were (0.90  0.22)
and (0.93  0.23) mL·min1, respectively. The AF of Tr
reached 57.37%. After Tr was orally administered, the Cmax
and Tmax values of (12.34  2.90) g·mL1 and (72.00  26.83)
min, respectively, were obtained. The results showed that Tr
was well absorbed, and could be mainly absorbed in the
small intestine [17], which would substantially explain its high
bioavailability. Tr was quickly eliminated, and its plasma
concentration fell by more than two-thirds within 30 min
after being intravenously administered, but dropped back to
the minimum value within 6 h after gavage or intravenous
administration of Tr.



Fig. 2 Mean plasma concentration–time profile of
Tr determined by HILIC-UPLC assay after intravenous
administration at 5 mg·kg1 and gavage administration
at 10 mg·kg1 in rats (n = 5)

The mean plasma concentration–time profile of Tr after
CHENG Zai-Xing, et al. /Chinese Journal of Natural Medicines 2013, 11(2): 164170
2013年 3月 第 11卷 第 2期 Chin J Nat Med Mar. 2013 Vol. 11 No. 2 169

MECB was orally administered is shown in Fig. 3. The main
pharmacokinetic parameters are reported in Table 3. The
AUC(0∞), t1/2, CL(s), Cmax, Tmax, and AF values obtained were
(4 566.75  1 435.64) min·g·mL1,, (225.76  25.28) min,
(0.75  0.18) mL·min1, (11.98  3.69) g·mL1, (60.00 
21.21) min, and 64.42%, respectively. The results indicated
that no significant difference was observed at the value of
AUC, Cmax, Tmax, CL(s) or t1/2 between the groups of Tr and
MECB orally administered (P > 0.05). Hence, other
ingredients had no affect on the bioavailability of Tr in
MECB.


Fig. 3 Mean plasma concentration–time profile of Tr in
MECB determined by HILIC-UPLC assay after gavage
administration at 10 mg·kg1 in rats (n = 5)

Table 3 Pharmacokinetic parameters of Tr in rats (mean  SD, n = 5)
Parameter Value (Tr, i.v.) Value (Tr, oral) Value (MECB, oral)
AUC(0∞)/( min·g·mL1) 3 544.29  908.80 4 066.83  1 244.41 4 566.75  1435.64
AUC(0t)/( min·g·mL1) 3 037.51  786.75 3 395.37  1 078.95 4 039.97  1 176.74
cmax/g·mL1 —— 12.34  2.90 11.98  3.69
Tmax/min —— 72.00  26.83 60.00  21.21
CL(s)/mL·min1 0.93  0.23 0.90  0.22 0.75  0.18
t1/2/min 201.49  37.33 215.86  15.91 225.76  25.28
MRT(0∞)/min 615.62  145.90 724.21  424.94 526.59  81.35
MRT(0t)/min 290.75  53.87 311.50  22.96 368.66  115.59
AF/% —— 57.37 64.42
RF/% —— —— 112.29
Data were obtained following intravenous administration of Tr at 5 mg·kg1, oral administration of Tr at 10 mg·kg1, and oral administration of Tr
in MECB at 10 mg·kg1. P < 0.05 were considered statistically significant

4 Conclusion
A simple, specific, and sensitive HILIC-UPLC assay was
developed and validated for the quantification of Tr in rat
plasma. The method exhibited excellent performance in terms
of high selectivity, low LLOQ (0.12 g·mL1), and wide
linear range (0.12–100 g·mL1). It was successfully used for
pharmacokinetic analysis of Tr, administered by gavage or
intravenously, the bioavailability of which was determined to
be good. Moreover, differences between Tr alone and Tr in
MECB orally administered were not detected.
Abbreviations
Tr, trigonelline; MECB, methanol extract of coffee beans;
HILIC, hydrophilic interaction liquid chromatography; UPLC,
ultra performance liquid chromatography; QC, quality
control; LLOQ, lower limit of quantitation; RSD, relative
standard deviation; AUC, area under the plasma concentra-
tion- time curve; cmax, peak plasma concentration; Tmax, peak
time; CL(s), clearance; t1/2, biological half-life; AF, absolute
bioavailability; RF, relative bioavailability; MRT, mean
residence time.
References
[1] Kimberly F, Katarina M, Jairam V, et al. Trigonelline is a novel
phytoestrogen in coffee beans [J]. J Nutr, 2009, 139(10): 1833-
1838.
[2] Ky CL, Guyot B, Louarn B, et al. Trigonelline inheritance in
the interspecific Coffea pseudozanguebariae×C. liberica var.
dewevrei cross [J]. Theor Appl Genet, 2001, 102(4): 630-634.
[3] Minamisawa M, Yoshida S, Takai N. Determination of
biologically active substances in roasted coffees using a
diode-array HPLC system [J]. Anal Sci, 2004, 20(2): 325-328.
[4] Sánchez-Hernández L, Puchalska P, García-Ruiz C, et al.
Determination of trigonelline in seeds and vegetable oils by
capillary electrophoresis as a novel marker for the detection of
adulterations in olive oils [J]. J Agric Food Chem, 2010, 58(13):
7489-7496.
[5] Shruti C, Farhan J, Roop K, et al. Validated high-performance
thin-layer chromatography method for determination of trig-
onelline in herbal extract and pharmaceutical dosage form [J].
Analyt Chim Acta, 2006, 577(1): 46-51.
[6] Hirakawa N, Okauchi R, Miura Y, et al. Anti-invasive activity
of niacin and trigonelline against cancer cells [J]. Biosci
Biotechnol Biochem, 2005, 69(3): 653-658.
[7] Carter MT, Kamau GN, Rusling JF. Electroreduction of
CHENG Zai-Xing, et al. /Chinese Journal of Natural Medicines 2013, 11(2): 16470
170 Chin J Nat Med Mar. 2013 Vol. 11 No. 2 2013年 3月 第 11卷 第 2期

pyridinium carboxylic acids homarine and trigonelline in acidic
solutions [J]. J Electr Chem Interfacial Electrochem, 1984,
170(1-2): 265-279.
[8] Zhao HQ, Qu Y, Wang XY, et al. Determination of trigonelline
by HPLC and study on its pharmacokinetics [J]. Acta Pharm
Sinica, 2003, 38(4): 279- 282.
[9] Zhao HQ, Qu Y, Wang XY, et al. Determination of trigonelline
in Trigonella foenum-graecum by HPLC [J]. China J China
Mater Med, 2002, 27(3): 194-196.
[10] Liu GX, Shang MY, Li H, et al. Extraction and determination
of trigonelline in Trigonella foenum-graecum seeds [J]. Drug
Stand China, 2005, 6(4): 11-14.
[11] Zhou RJ, Wang L, Wang LX, et al. Determination of
trigonelline in Trigonella foenum graecum L. by hydrophilic
interaction chromatography [J]. Chin J Chromatogr, 2010,
28(4): 379-382.
[12] Cheng ZX, Yan TM, Chen H, et al. Determination of
trigonelline in Pinellia ternata by UPLC [J]. Chin J Exp Trad
Med Form, 2012, 18(4): 85-87.
[13] Gu JF, Chen NZ, Ding GQ, et al. Determination of oxaceprol
in rat plasma by LC-MS/MS and its application in a
pharmacokinetic study [J]. J Pharm Biomed Anal, 2011, 54(1):
173-178.
[14] Chen LL, Yuan WW, Hu ZF, et al. Determination and
pharmacokinetics of DT-13 in rat plasma by LC-MS [J]. J
Pharm Biomed Anal, 2011, 56(3): 650-654.
[15] Li HZ. Biopharmaceutical analysis [M]. Beijing: China Press
of Traditional Chinese Medicine, 2004: 131.

建立 HILIC-UPLC联用分析法测定血浆中葫芦巴碱及在药动学研
究中的应用
程再兴 1, 4，吴锦俊 3，刘中秋 2*，林 娜 1, 4*
1福建中医药大学药学院，福州 350108；
2南方医科大学药学院，广州 510515；
3广州中医药大学中药学院，广州 510006；
4中国中医科学院中药研究所，北京 100007
【摘 要】 目的：葫芦巴碱是咖啡豆中含量较多的生物碱之一，为了研究其药代动力学特征，本研究建立了一种简便、灵
敏、准确的亲水相互作用色谱超高效液相联用分析方法，并研究了葫芦巴碱和咖啡豆甲醇提取物的药代动力学。方法：大鼠灌
胃和尾静脉注射给予葫芦巴碱和咖啡豆甲醇提取物后，眼眶静脉丛取血，血液样本采用亲水相互作用色谱超高效液相联用分析
法进行分析，药代动力学参数按非房室模型，通过药动学软件 87/97 进行计算。结果：对所建立的方法进行了方法学的考察，
线性范围为 0.12–100 g·mL1，最低定量下限为 0.12 g·mL1，日内日间精密度、准确度、提取回收率和稳定性等均达到体内药
物分析要求。葫芦巴碱口服和尾静脉注射给药后，AUC(0∞)分别为(4 066.83  1 244.41)和(3 544.29  908.80) min·g·mL1，生物
利用度达到 57.37%。咖啡豆甲醇提取物口服后，AUC(0-∞)为(4 566.75  1 435.64) min· g·mL1，绝对生物利用度为 64.42%，相对
生物利用度为 112.29%。结论：所建立的亲水作用色谱超高效液相联用分析法简单、准确、重现性好；葫芦巴碱无论是纯品还
是在咖啡豆甲醇提取物中都有较好的生物利用度，而且没有明显的差异。
【关键词】 亲水相互作用色谱-超高效液相联用分析法；葫芦巴碱；咖啡豆；生物利用度

【基金项目】 国家重点基础研究发展计划（973计划）(Nos. 2009CB5228008, 2011CB505300, 2011CB505305)和国家自然基金项
目(No. U0832002)资助