Research Article

Received: 28 October 2010;

Revised: 25 March 2011;

Accepted: 21 April 2011

Published online in Wiley Online Library: 2 August 2011

(wileyonlinelibrary.com) DOI 10.1002/pca.1335

HPLC Analysis of Oxindole Alkaloids in Uncaria

Tomentosa: Sample Preparation and Analysis
Optimisation by Factorial Design
Gustavo Bertol, Luzia Franco and Brs Heleno de Oliveira*
ABSTRACT:
Introduction Uncaria tomentosa (cats claw) is widely used for the treatment of some infectious and inammatory
diseases. Oxindole alkaloids are regarded as the most important components responsible for the biological activities
attributed to the plant. Their analysis require efcient sample preparation and suitable reference standards but few are
commercially available.
Objective To develop and validate a HPLC analytical method for oxindole alkaloids in Uncaria tomentosa with emphasis on
sample preparation.
Method Factorial experimental designs were used for the optimisation of both sample preparation and chromatographic
separation. The optimised sample preparation involved extraction with aqueous ethanol, and the granulometry of the
powdered plant material signicantly inuenced extraction yields. Mitraphylline was used as a calibration reference for the
determination of total alkaloids.
Results The method was fully validated and showed good selectivity, linearity (r2 0.9996), accuracy ( 96%) and precision
(RSD < 2.4%). Detection and quantication limits for mitraphylline were 0.8 and 2.4 ppm, respectively.
Conclusion The optimised chromatographic method, using organic buffer in the mobile phase, provided baseline separation
of tetracyclic and pentacyclic alkaloids in the samples. Calibration using mitraphylline provided more accurate estimates of
total alkaloid content when compared to other available reference alkaloids. Copyright 2011 John Wiley & Sons Ltd.
Keywords: Factorial experimental design; HPLC; alkaloids; Uncaria tomentosa

Introduction

Phytochem. Anal. 2012, 23, 143151

* Correspondence to: B. H. de Oliveira, Universidade Federal do Paran,

Departamento de Qumica, CP19081, CEP 81531990, CuritibaPR, Brazil.
Email: bho@ufpr.br
Universidade Federal do Paran, Departamento de Qumica, CP19081, CEP
81531990 CuritibaPR, Brazil

Copyright 2011 John Wiley & Sons, Ltd.

143

Uncaria tomentosa (Willd.) DC is a vine known as cats claw that

is used in traditional medicine by natives of South and Central
America. The plant is now used worldwide in folk and
complementary medicine for the treatment of some infectious
and inammatory diseases, and it also has hypotensive and
immunomodulatory properties (Keplinger et al., 1999; Heitzman
et al., 2005). Toxicological studies have also indicated that
traditional preparations and uses of cats claw are safe (Valerio
and Gonzales, 2005).
Among the various classes of chemical constituents found in
the plant, oxindole alkaloids are regarded as the most important
(Laus et al., 1997). Two subtypes of oxindole alkaloids, pentacyclic
(POA) and tetracyclic (TOA), have been found in U. tomentosa
(Fig. 1) and their prole varies seasonally and geographically.
Moreover, under certain conditions they are interconvertible
(Laus et al., 1996), which should be considered during sample
preparation for analysis.
The alkaloid prole in U. tomentosa is important because POA
and TOA have distinct pharmacological properties. TOA (for
example 7 and 8) have been shown to be active on cardiovascular
and central nervous systems (Shi et al., 2003), whereas POA are
associated with immunostimulatory properties (Heitzman et al.,
2005). Additionally, besides being nonimmunostimulants
(Keplinger et al., 1999), TOA have an antagonistic effect against
POA (Wurm et al., 1998). Due to these important pharmacological

differences the United States Pharmacopoeia (USP) has established the limit of 0.05% of TOA relative to POA in products
containing the plant (USP, 2008). Therefore, analytical procedures
for quality control of plant material or phytomedicines containing
U. tomentosa should be able to distinguish these alkaloid classes
and avoid their interconversion during sample preparation.
A number of HPLC procedures for alkaloid determination in
U. tomentosa have been described (Stuppner et al., 1992; Laus
and Keplinger, 1994; Ganzera et al., 2001). Most of them use
silica C18columns but their selectivity, and even elution order,
may change from one manufacturer to another, requiring
changes in the conditions. Mobile phases containing inorganic
buffers at different pH values are often used. Although efcient,
these inorganic buffers may reduce column life and increase
overall analysis time due to the long postanalysis column
washing time to avoid salt precipitation. The use of organic
buffers, such as the volatile triethylammonium acetate, avoids
these shortcomings.
Most reports, and a USP monograph, use pteropodine as a
calibration reference for determination of the total alkaloids in

G. Bertol et al.

Figure 1. Structures of oxindole alkaloids from Uncaria tomentosa: speciophylline (1), uncarine F (2), mitraphylline (3), isomitraphylline (4),
pteropodine (5), isopteropodine (6), rhynchophylline (7) and isorhynchophylline (8).

144

U. tomentosa (Laus and Keplinger, 1994; Ganzera et al., 2001;

USP, 2008). This is because not all reference alkaloids from the
plant are commercially available and because their chromophores are similar. However, no comparative study was found in
order to determine the most suitable reference compound for
that purpose.
Statistical factorial design is a technique that allows the
experimenter to simultaneously evaluate multiple variables that
affect a particular process, instead of one at a time. The main
advantage is that the optimisation may be achieved quickly
(Montgomery, 2001). It has been successfully used in chromatography for optimising both sample preparation and analytical
separation (Ferreira et al., 2007).
The objective of this work was to develop a HPLC method for
determination of total POA in U. tomentosa, using mitraphylline
as a calibration reference, and to make comparisons with
other alkaloid markers from the plant. The method, using
triethylammonium acetate buffer, should be capable of
distinguishing the different alkaloid patterns of this species.
Sample preparation was designed to maximise recovery and
reduce alkaloid isomerisation. In order to achieve these goals,
multivariate statistical experimental designs were used for the
optimisation of both plant extraction and HPLC separations.

wileyonlinelibrary.com/journal/pca

Experimental
Instrumentation
HPLC analyses were performed with an 1100 series (Agilent Technologies, Torrance, USA), equipped with degasser (G1379A), quaternary
pump (G1311A), autosampler (G1313A), column oven (G1316) and
photo diode array (PDA) detector (G1315B). Mass spectra were recorded
with an atmospheric pressure ionization (API) 3000 spectrometer
(Applied Biosystems/MDS Sciex, CA. USA), with an electrospray ion
source in positive mode (ESI+). Data acquisition and processing were
carried out using the software Analyst version 1.4.1. All samples (5 mg)
for MS analysis were dissolved in ethanol, 5 mM ammonium acetate
(9:1 v/v, 10 mL). The interface was maintained at 500 C, ultrapure
synthetic air was used as nebulisation and desolvation gas (6 psi) and
the curtain gas was nitrogen (6 psi). The ionisation potential was set at
5.5 kV and declustering and focalisation potentials were set at 50 and
200 V respectively. NMR spectra were acquired with a DRX400
instrument (Bruker, Rheinstetten, Germany), with samples prepared in
CDCl3 using trimethylsilyl (TMS) as an internal standard.

Plant material
Two samples of U. tomentosa were used for comparison. The rst (UtB)
was collected in the state of Acre, Brazil and a voucher specimen was

Copyright 2011 John Wiley & Sons, Ltd.

Phytochem. Anal. 2012, 23, 143151

Uncaria Tomentosa HPLC Analysis

deposited at Instituto de Botnica de So Paulo (#SP373337). Another
sample (UtP) was purchased from Peruvian Heritage (Lima, Peru) and
identied at the Museo de Historia Natural (#47USM2008).

The 22 full factorial design with four axial points and centre point is
summarised in Table 2.

Optimised conditions for HPLC analysis

Chemicals
Isomitraphylline (batch: 09417101; purity: 80.5%), pteropodine (batch:
21220101; purity: 99.9%) and isopteropodine (batch: 21225101; purity:
98.6%) were purchased from Chromadex (Santa Ana, USA). Acetic acid
and acetonitrile were purchased from J.T. Baker (USA). Ethanol, methanol,
sodium phosphate dibasic and ammonium hydroxide were supplied by
Vetec (Brazil). Triethylamine was purchased from Fluka (Germany).
Potassium dihydrogen phosphate was purchased from Merck (Germany)
and Polyamide 6 from MachereyNagel (Germany). Mitraphylline was
isolated and puried by column chromatography on silicagel (Shellard
et al., 1967); ppectrometric analysis data (MS, 1H and 13CNMR) were
compatible with its structure and literature data (Toure et al., 1992).

Statistical experimental designs for optimisation of

HPLC analysis
Factorial experimental designs and analysis were performed using
Minitab 15 statistical software (Minitab Inc. 2006, State College, PA, USA).
A 23 full factorial design with a complete and randomised repetition (16
runs) was generated for the screening of signicant conditions that
affect the separation of the eight alkaloids. The response was the
resolution of peaks of mitraphylline (3) and pteropodine (5) (Fig. 2). The
three factors chosen for evaluation and their respective high and low
levels are summarised in Table 1.
After screening, a central composite design (CCD) was set up for
resolution optimisation using the statistically signicant factors previously determined from screening (buffer pH and column temperature).

All analyses were made with a Zorbax XDB C18column (150 mm 4.6
mm, 3.5 m). The mobile phase was 35 mM triethylammonium acetate
buffer (0.2% v/v), pH 6.9 (A) and acetonitrile (B). Gradient composition (A:
B): 018 min (65:35); 1832 min (50:50). Flow rate was 0.8 mL/min and
detection was at 245 nm. Peaks were assigned on the basis of their
retention times compared to standards, when available, and by MS
analysis monitoring molecular ion [M + H]+ at m/z 369 for POA and m/
z 385 for TOA. For quantication of all POA a vepoint calibration curve
of mitraphylline was used (8200 g/mL).

Experimental design for optimisation of sample preparation

The dried bark was milled and sieved. The powder was weighed
(200 mg), transferred to a 10 mL volumetric ask, the volume was
completed with aqueous ethanol and the mixture sonicated. An aliquot
was then cleaned up by solid phase extraction (SPE) in polyamide
(200 mg) before analysis. The screening of signicant factors was made
with a 25 factorial design matrix with duplicates, which resulted in 64
experiments. The ve factors and their respective levels are shown in
Table 3 and the response was total alkaloid yield.
Optimisation was carried out using a central composite design using
the signicant parameters determined in the screening step (time,
solvent and temperature). The factors and respective levels in the design
are summarised in Table 4. Particle size was kept at 125180 m, sample
mass at 200 mg and solvent volume at 10 mL. The response was total
alkaloid yield.

Phytochem. Anal. 2012, 23, 143151

Copyright 2011 John Wiley & Sons, Ltd.

wileyonlinelibrary.com/journal/pca

145

Figure 2. Comparison of HPLC chromatograms of Uncaria tomentosa bark extracts according to sample origin. (A) UtB (Brazil) before separation
optimisation; (B) UtB after separation optimisation; (C) UtP (Peru) after separation optimisation. Key: speciophylline (1), uncarine F (2), mitraphylline (3),
isomitraphylline (4), pteropodine (5), isopteropodine (6), rhynchophylline (7) and isorhynchophylline (8).

G. Bertol et al.
Table 1. Factors and respective low and high values used
for screening of signicant variables for HPLC separation of
Uncaria tomentosa alkaloids

Table 3. Factors and respective low and high values used

for screening of signicant variables for extraction of
alkaloids from Uncaria tomentosa bark

Factor

Low

High

Factor

(A) Buffer
(B) pH
(C) Column temperature (C)

TEA
6
15

KH2PO4
6.9
35

(A) Time (min)

(B) Ethanol in water (%)
(C) Particle size (m)
(D) Temperature (C)
(E) Mass (mg)

TEA = triethylammonium acetate.

Optimised sample preparation conditions

Low
10
50
250355
25
100

High
30
90
125180
45
200

210.0 g mL1) were prepared in ethanol and injected on three

consecutive days into the HPLC system. Plant material (100, 150,
200, 250 and 300 mg) was extracted as described above and
analysed. Data were then statistically evaluated for r, R2 and
residuals of regression analysis.

The dried bark powder (particle size 125180 m) was weighed (200 mg)
and extracted with 60% ethanol (ca. 8 mL) in an ultrasonic bath for
20 min at 30 C. The volume was completed to 10 mL with the same
solvent. An aliquot (2 mL) was then cleaned up by SPE on polyamide
(200 mg) before analysis. The cartridge was conditioned with methanol
(2 mL) and then water (5 mL). After sample introduction the alkaloids
were immediately eluted with 60% ethanol (2 mL), the eluate was
transferred to a 5 mL volumetric ask and the volume completed with
the same solvent.

Accuracy
The true alkaloid concentration was determined using an
integrated calibration method (Miller, 1991). Thus a regular
calibration plot was made with standard mitraphylline (3)
solutions (8.0, 20.0, 40.0, 83.0 and 210.0 g mL1). Then, a
standard addition calibration curve was made with plant
material (200 mg) spiked with 1, 2, 3 and 4 mL of mitraphylline
solution (210 g mL1). After extraction, as described above, the
samples were analysed and the data were plotted in the same
graph. The concentration of alkaloid was determined by
extrapolating the value of the linear coefcient of the equation
for spiked sample solutions into that of the standard equation,
and the concentration was corrected using a factor calculated
for dilution and sample size. From the true concentration values
thus obtained, samples with 50, 100 and 150% of theoretical
values were prepared in triplicate, under the same conditions
used for linearity tests. After analysis recovery was calculated.

Comparison of reference alkaloids

Solutions of mitraphylline, isomitraphylline, pteropodine and isopteropodine were prepared in ethanol (8.0, 20.0, 40.0, 83.0 and 210.0 g/mL,
the respective analytical curves were constructed and absorptivity was
calculated at all concentrations. Bark samples (n = 3) were chromatographed and those peaks corresponding to commercially available
standards were quantitated against their specic calibration curves. Then
all peaks in the chromatograms, including those for the commercially
unavailable speciophylline and uncarine F, were quantitated using a
single calibration curve of each of the four standards. The results were
then compared (ANOVA, Tukeys test).

Method validation
Selectivity

Precision

Selectivity was evaluated by the peak purity function of the

chromatography software and by direct inspection of the UV
spectra at upslope, apex and downslope portions of the peak
of analytes in HPLCPDA chromatograms. Matrix effect was
assessed by comparing a calibration graph prepared with the
standard with that prepared with plant extract samples spiked
with the standard, as described below (Accuracy).

Withinday precision (repeatability) was evaluated by repeated

analyses of plant material (n = 6), and betweenday precision
(intermediate precision) was evaluated by two analysts on two
consecutive days (n = 6). The concentration of total alkaloids
against the mitraphylline standard curve was determined and
the relative standard deviation (RSD) calculated and compared.

Linearity

Detection and quantitation limits

Linearity was determined for standard and sample solutions.

Mitraphylline standard solutions (8.0, 20.0, 40.0, 83.0 and

Detection and quantitation limits (LOD and LOQ respectively)

were calculated according to the equations LOD = 3.3/S and

Table 2. Central composite design factors and respective values used for optimisation of HPLC separation of Uncaria tomentosa
alkaloids
Experimental domaina

Factor

Buffer pH
Column temperature (C)

6.5
10

6.8
12.2

7.5
17.5

8.2
22.8

8.5
25

Design points: centre (0), cubic (1, 1), axial (, ); n = 5 at the centre.

146

wileyonlinelibrary.com/journal/pca

Copyright 2011 John Wiley & Sons, Ltd.

Phytochem. Anal. 2012, 23, 143151

Uncaria Tomentosa HPLC Analysis

Table 4. Central composite design factors and levels for optimisation of extraction of alkaloids from Uncaria tomentosa bark
Experimental domaina

Factor

Time (min)
Ethanol in water (%)
Temperature (C)

20
30
30

22
38
33

25
50
38

28
62
42

30
70
45

Design points: centre (0), cubic (1, 1), axial (, ).

LOQ = 10/S, where is the standard deviation of responses and S

is the slope of the analytical curve (Toure et al., 1992). Appropriate
dilutions were then prepared and analysed in order to conrm
those limits.

Results and Discussion

Statistical multivariate designs have been used increasingly for
analytical purposes. Many variables can affect sample preparation
and chromatographic separation and the possibility of evaluating
many factors at once, instead of one a time, is cost and time
effective. The procedure usually involves an initial screening step
in order to identify statistically signicant factors followed by an
optimisation step using a surface response method (Ferreira et al.,
2007). This twostep methodology was used for the optimisation
of alkaloid separation by HPLC and for sample preparation.

HPLC optimisation
Initial attempts at HPLC separation of the eight alkaloids present
in sample UtB using a literature procedure (Ganzera et al., 2001)
was not successful (Fig. 2A). The peaks of isomitraphylline (4)

and pteropodine (5) coeluted and those of uncarine F (2) and

mitraphylline (3) showed poor separation. The optimisation of
chromatographic separation was performed in two steps.
Initially, screening experiments were carried out in order to
determine the signicant factors for improved separation. The
factors and respective levels in the experimental design are
summarised in Table 1. The responses were resolution of peaks
3 and 5 of the critical pairs 23 and 45. Resolution (R) was
calculated using retention time (tr) and peak width at baseline
(twb) according to the formula: R = [2(tr2 tr1)]/(twb1 + twb2). The
results obtained for each run are listed in Table 5.
After calculation of the main effects and interactions the
statistically signicant factors could be determined. The graphic
representations of the results for pterodine (5) (Fig. 3) was
similar to that of mitraphylline (3) (not shown). The plot of main
effects indicated a more intense effect of temperature on
resolution. The Pareto chart showed that the most signicant
factor was column temperature followed by a much smaller
effect for pH. The effect of temperature was negative, i.e.,
increasing temperature reduces resolution. Many studies have
shown that resolution can be improved at higher temperatures
but the relationship is not always linear. The separation of the
diastereomers of an experimental serine protease inhibitor on

Table 5. Full factorial matrix for screening of HPLC conditions for alkaloid separation and resolution results of individual
experiments
Run order

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16

Factor
Buffer*

pH

TEA
TEA
TEA
TEA
KH2PO4
KH2PO4
KH2PO4
KH2PO4
KH2PO4
TEA
KH2PO4
TEA
KH2PO4
TEA
KH2PO4
TEA

7
7
6
6
7
6
6
6
7
7
7
6
7
6
6
7

Resolution (R)
Temperature (C)
15
15
35
35
15
35
35
15
35
35
35
15
15
15
15
35

Mitraphylline (3)

Pterodine (5)

2.86
2.80
0.00
0.00
3.74
0.00
0.00
0.00
2.44
1.55
2.20
1.64
3.84
1.76
0.00
1.54

2.09
2.10
0.00
0.00
2.19
0.00
0.00
1.68
0.57
0.00
0.00
1.59
2.34
1.59
1.80
0.00

Phytochem. Anal. 2012, 23, 143151

147

TEA = triethylammonium acetate.

Copyright 2011 John Wiley & Sons, Ltd.

wileyonlinelibrary.com/journal/pca

G. Bertol et al.

Figure 3. Main effects plot and Pareto chart from results of screening of signicant effects for the HPLC resolution of pterodine (5).

C18columns showed that, in the range of 2545 C, resolution

was lower than that at higher temperatures (Cunliffe et al.,
2011). That also happened in our case in the range studied. The
effect of pH, on the other hand, was positive because at higher
pH the resolution increased.
These preliminary results also showed that type of buffer
(organic or inorganic) did not statistically affect the resolution.
Therefore, organic buffer was chosen for all subsequent
experiments.
The next step, the optimisation of pH and temperature, was
carried out using a CCD. The design matrix and results are
summarised in Table 6. The optimal conditions could be inferred
from the response surfaces for peaks 3 and 5 (Fig. 4). The inuence
of pH was rationalised in terms of the alkaloids pKas. Considering
that the pKas for all POA are between 5.2 and 6.7 (Laus et al., 1996),
a pH higher than 6.8 ensures that N4 is not protonated, which
leads to a better columnalkaloid interaction and thereby
improving chromatographic separation. Values of pH above 7.0
lead to a coelution of rhynchophylline (7) with isomitraphylline
(4). Therefore, pH was set at 6.9 in the nal method.
Good resolutions for both critical peak pairs were achieved
with column temperature at 15 C. A chromatogram obtained
under these conditions shows baseline separation (resolution
1.5) and good symmetry for all peaks (Fig. 2C). The elution order

was the same as in previous reports (Laus and Keplinger, 1994;

Ganzera et al., 2001).
Plant extraction optimisation
Statistical experimental design was also used for the optimisation of sample extraction. Five factors (time, solvent composition, particle size, temperature and sample mass) were selected
for initial screening of the statistically signicant variables
(Table 3). The response was total alkaloid yield calculated using
mitraphylline as reference. The results showed that extraction
time, percentage ethanol in water, particle size and temperature
were statistically signicant (Fig. 5).
The most important factor was granulometry. A decrease by half
in particle size (from 250355 to 125180 m) led to a very large
increase in extraction yield. Therefore, particle size was kept in the
range 125180 m for the subsequent optimisation experiments.
The next step, the optimisation of the signicant parameters,
was carried out using CCD. The factors and respective levels in
the experimental design are summarised in Table 4. The results,
shown in Fig. 6, indicate that the best concentration of ethanol
in the extraction solvent was in the range 6070%. Optimum
extraction time was 20 min at 30 C. These conditions were then
used for method validation and sample analysis.

Table 6. Central composite design matrix and results for HPLC resolution optimisation of mitraphylline (3) and pteropodine (5)
Run

148

1
2
3
4
5
6
7
8
9
10
11
12
13

pH

Temperature (C)

Mitraphylline (R)

Pteropodine (R)

8.50
7.50
7.50
8.21
7.50
7.50
7.50
6.79
7.50
6.79
6.50
8.21
7.50

17.50
17.50
17.50
22.80
10.00
17.50
25.00
22.80
17.50
12.20
17.50
12.20
17.50

4.11
3.67
3.71
3.60
4.11
3.75
3.10
2.52
3.68
3.21
1.87
4.09
3.56

2.15
2.08
2.09
1.59
2.84
2.09
1.35
1.43
2.09
2.55
1.86
2.64
2.06

wileyonlinelibrary.com/journal/pca

Copyright 2011 John Wiley & Sons, Ltd.

Phytochem. Anal. 2012, 23, 143151

Uncaria Tomentosa HPLC Analysis

Figure 4. Response surface plot for the resolution of (a) mitraphylline (3) and (b) pteropodine (5).

Figure 5. Main effects plot and Pareto chart for the screening of parameters for extraction of alkaloids from the bark of Uncaria tomentosa.

The most common sample preparation procedures described

for U. tomentosa involve extraction with methanol or ethanol
(Stuppner et al., 1992; Laus and Keplinger, 1994; Ganzera et al.,
2001), but a supercritical uid extraction has also been used
successfully (LopezAvila et al., 1997). The optimised extraction
conditions developed in this study were compared with two
others commonly used methods. The rst was a single
extraction with methanol, and the other was multiple extractions with 60% ethanol. The experiments were performed in
triplicate and the means were compared. The results (Table 7)
showed that the optimised extraction procedure described
provides a small but statistically signicant improvement in
alkaloid extraction.
Comparison of reference alkaloids

Phytochem. Anal. 2012, 23, 143151

Method validation
The method was validated according to current guidelines of
the International Conference on Harmonisation (ICH, 2005) and
Sanitary Vigilance National Agency (ANVISA, 2003). Selectivity
was evaluated by the spectral purity of POA peaks. The peak
purity function of the HPLC software used indicated no co
eluting peaks, which was conrmed by more detailed inspection
of UV data at different positions of each peak (upslope, apex

Copyright 2011 John Wiley & Sons, Ltd.

wileyonlinelibrary.com/journal/pca

149

Pteropodine is commonly used as a calibration reference for

determination of the total alkaloid content in U. tomentosa.
In order to determine the most suitable general calibration
reference, alkaloid solutions of standard mitraphylline,

isomitraphylline, pteropodine and isopteropodine were analysed

and the concentration:area ratio was calculated for all peaks in
all concentrations used (Table 8). Comparison of the means
(ANOVA) showed similar absorptivity for alkaloids with the
same conguration at C7 (mitraphylline/pteropodine and
isomitraphylline/isopteropodine). The differences from the
control, although small, were signicant for all alkaloids except
for mitraphylline. This alkaloid, therefore, is the most suitable
reference for the determination of total alkaloids in the plant.
(Table 9)

G. Bertol et al.
Table 7. Comparison of extraction conditions of alkaloids
from Uncaria tomentosa. (A) Control, optimised conditions
from this work; (B) same as A but using methanol as
extracting solvent; (C) same as A but with three extractions
Conditions

POA (%)
(mean SD)

A
B*
C*

1.03 0.1
0.93 0.1
1.00 0.4

*p < 0.05.

Table 8. Comparison of absorptivity for calibration solutions mitraphylline, isomitraphylline, pteropodine and isopteropodine
Alkaloid

Concentration range
(g mL1)

Mitraphylline
Isomitraphylline
Pteropodine
Isopteropodine

8.10203.00
6.50163.0
8.60214.00
8.10203.00

Concentration/
area
0.0334 0.0002
0.0359 0.0003
0.0322 0.0004
0.0359 0.0004

Table 9. Comparison of total alkaloid content of Uncaria

tomentosa bark according to reference alkaloid used. Control
is the regular calibration using all analytical curves.
Speciophylline was calculated as isopteropodine and uncarine F as mitraphylline
Reference
Mitraphylline
Isomitraphylline
Pteropodine
Isopteropodine
Control

Total alkaloid (%)

1.07 0.02
1.15 0.02*
1.04 0.01*
1.15 0.02*
1.09 0.02

*p < 0.05, n = 3.

and downslope). Other validation parameters are summarised in

Table 10.
Application to real samples
The optimised and validated method was then used for the
quantication of alkaloids (as mitraphylline) in U. tomentosa
bark samples. The Brazilian plant sample (UtB) contained POA
only (1.05 0.02%), whereas the Peruvian sample (UtP)
contained POA (0.88 0.02%) and TOA (0.31 0.02%). The latter,
according to the USP, is not suitable for use in phytomedicines
due to its high content of TOA (USP, 2008).

150

Figure 6. Response surface plot for optimisation of extraction of

alkaloids from the bark of Uncaria tomentosa.

wileyonlinelibrary.com/journal/pca

Copyright 2011 John Wiley & Sons, Ltd.

Phytochem. Anal. 2012, 23, 143151

Uncaria Tomentosa HPLC Analysis

Table 10. Method validation data for alkaloids determination in Uncaria tomentosa

Linearity (standard)

Linearity (sample)

Precision
Accuracy

Detection limit
Quantitation limit

Parameter

Results

Function
R
R2
Residuals (homocedasticity)
Function
R
R2
Residuals (homocedasticity)
Intraday
Intermediate
Recovery (50% theoretical concentration)
Recovery (100% theoretical concentration)
Recovery (150% theoretical concentration)
LOD
LOQ

y = 31.108x 2.5938
0.9998
0.9996
Aleatory distribution
y = 30.333x 297.97
0.9979
0.9959
Aleatory distribution
RSD < 1.45% (n = 6)
RSD < 2.29% (n = 12)
98.43% (n = 3)
98.37% (n = 3)
98.89% (n = 3)
0.80 ppm
2.4 ppm

Acknowledgements
The nancial support of CNPq and Fundao Araucria (Brazil) is
gratefully acknowledged.

References
ANVISA. 2003. ResoluoRE n. 899, de 29 de maio de 2003. Dispe sobre
o Guia para validao de mtodos analticos e bioanalticos. Agncia
Nacional de Vigilncia Sanitria (http://www.anvisa.gov.br/eng/index.
htm).
Cunliffe JM, Dreyer DP, Hayes RN, Clement RP, Shen JX. 2011. Using
temperature to optimize resolution and reduce analysis times for
bioanalytical diastereomer LCMS/MS separations. J Pharmaceut
Biomed Anal 54: 179185.
Ferreira SLC, Bruns RE, da Silva EGP, dos Santos WNL, Quintella CM,
David JM, de Andrade JB, Breitkreitz MC, Jardim ICSF, Neto BB. 2007.
Statistical designs and response surface techniques for the
optimization of chromatographic systems. J Chromatogr A 1158:
214.
Ganzera M, Muhammad I, Khan RA, Khan IA. 2001. Improved method for
the determination of oxindole alkaloids in Uncaria tomentosa by high
performance liquid chromatography. Planta Med 67: 447450.
Heitzman ME, Neto CC, Winiarz E, Vaisberg AJ, Hammond GB. 2005.
Ethnobotany, phytochemistry and pharmacology of Uncaria
(Rubiaceae). Phytochemistry 66: 529.
ICH. 2005. Validation of Analytical Procedures: Text and Methodology. ICH
Guidelines, Q2 (R1), International Conference on Harmonisation:
Geneva (www.emea.europa.eu/pdfs/human/ich/038195en.pdf).
Keplinger K, Laus G, Wurm M, Dierich MP, Teppner H. 1999. Uncaria
tomentosa (Willd.) DC Ethnomedicinal use and new pharmacological, toxicological and botanical results. J Ethnopharmacol 64:
2334.

Laus G, Keplinger D. 1994. Separation of stereoisomeric oxindole

alkaloids from Uncaria tomentosa by highperformance liquid
chromatography. J Chromatogr A 662: 243249.
Laus G, Brossner D, Senn G, Wurst K. 1996. Analysis of the kinetics of
isomerization of spiro oxindole alkaloids. J Chem Soc Perk 2 9: 19311936.
Laus G, Brossner D, Keplinger K. 1997. Alkaloids of Peruvian Uncaria
tomentosa. Phytochemistry 45: 855860.
LopezAvila V, Benedicto J, Robaugh D. 1997. Supercritical uid
extraction of oxindole alkaloids from Uncaria tomentosa. HRC J High
Resolut Chromatogr 20: 231236.
Miller JN. 1991. Basic statistical methods for analytical chemistry. 2.
Calibration and regression methods a review. The Analyst 116: 314.
Montgomery DC. 2001. Design and Analysis of Experiments, 5th edn. John
Wiley & Sons: New York.
Shellard EJ, Beckett AH, Tantivat P, Phillips JD, Lee CM. 1967. Mitragyna
species of Asia. 8. Alkaloids of leaves of Mitragyna javanica Var
Microphylla Koord and Valeton. Planta Med 15: 245248.
Shi JS, Yu JX, Chen XP, Xu RX. 2003. Pharmacological actions of Uncaria
alkaloids, rhynchophylline and isorhynchophylline. Acta Pharmacol
Sin 24: 97101.
Stuppner H, Sturm S, Konwalinka G. 1992. HPLC analysis of the main
oxindole alkaloids from Uncaria tomentosa. Chromatographia 34:
597600.
Toure H, Babadjamian A, Balansard G, Faure R, Houghton PJ. 1992.
Complete H1 and C13 NMR chemicalshift assignments for some
pentacyclic oxindole alkaloids. Spectrosc Lett 25: 293300.
USP. 2008. United States Pharmacopeia, Rockville, MD (www.USP.org).
Valerio J, Gonzales GF. 2005. Toxicological aspects of the South
American herbs cats claw (Uncaria tomentosa) and maca (Lepidium
meyenii): A critical synopsis. Toxicol Rev 24: 1135.
Wurm M, Kacani L, Laus G, Keplinger K, Dierich MP. 1998. Pentacyclic
oxindole alkaloids from Uncaria tomentosa induce human endothelial cells to release a lymphocyteproliferationregulating factor.
Planta Med 64: 701704.

151

Phytochem. Anal. 2012, 23, 143151

Copyright 2011 John Wiley & Sons, Ltd.

wileyonlinelibrary.com/journal/pca