Journal of Ethnopharmacology 148 (2013) 37–44

Contents lists available at SciVerse ScienceDirect

Journal of Ethnopharmacology
journal homepage: www.elsevier.com/locate/jep

Identification of active compounds from Caesalpinia sappan L. extracts

suppressing IL-6 production in RAW 264.7 cells by PLS
Ming-Juan Chu a,1, You-Zhi Wang a,1, Kiyoshi Itagaki b, Hong-Xing Ma c, Ping Xin a,
Xue-Gang Zhou a, Guo-You Chen a, Sen Li a, Shi-Qin Sun a,n
a
Department of Biopharmaceutical Sciences, College of Pharmacy, Harbin Medical University-Daqing, Daqing 163319, China
b
Department of Surgery, Beth Israel Deaconess Medical Center and Harvard Medical School, Boston 02215, USA
c
Clinical Lab of Daqing Oil General Hospital, Daqing 163001, China

art ic l e i nf o a b s t r a c t

Article history: Ethnopharmacological relevance: Caesalpinia sappan L. is distributed in Southeast Asia and also used as
Received 22 October 2012 herbal medicine for the treatment of various diseases such as burning sensations, leprosy, dysentery,
Received in revised form osteoarthritis and rheumatoid arthritis (RA). The overproduction of IL-6 plays an important role in the
14 March 2013
prognosis of RA, but the active compounds from the extracts of Caesalpinia sappan L. suppressing IL-6
Accepted 18 March 2013
Available online 6 April 2013
production remain unknown.
Aims of the study: Identifying the main active compounds of Caesalpinia sappan L. extracts inhibiting the IL-6
Keywords: production in LPS-stimulated RAW 264.7 cells by partial least squares (PLS).
Caesalpinia sappan L. Materials and methods: Sixty-four samples with different proportions of compounds were prepared
RAW 264.7
from Caesalpinia sappan L. by supercritical CO2 fluid extraction (SCFE) and refluxing. Each of 64 samples
IL-6
was applied to RAW 264.7 cells with LPS to evaluate whether IL-6 production by LPS is affected by addition
PLS
Active compounds of each sample. The IL-6 production in medium was determined by ELISA and the inhibitory activity of each
sample was analyzed. In addition, the fingerprints of these 64 samples were also established by ultra-
performance liquid chromatography electrospray ionization tandem mass spectrometry (UPLC–MS). We
used the PLS, a simplified method, to evaluate the results from IL-6 production and fingerprints.
Results: Each of 64 samples markedly suppressed LPS-induced IL-6 production in RAW cells. The
fingerprints by UPLC–MS clearly revealed variations among 64 samples produced in different extract
conditions. The PLS analysis with IL-6 production and fingerprints by UPLC–MS suggested that the peaks 71,
93, 150, 157, 168 have more influence on the inhibitory activity of Caesalpinia sappan L. extracts. The peaks
71, 93, 150 are likely representing sappanone A, protosappanin E and neoprotosappanin, respectively. The
peaks 157 and 168 are still at large.
Conclusion: This is the first report that sappanone A, protosappanin E, neoprotosappanin and two
unidentified compounds can be considered as possible active compounds that might inhibit IL-6 production.
Further studies are needed to confirm the effectiveness of these five compounds on IL-6 production and
possible mechanism.
& 2013 Elsevier Ireland Ltd. All rights reserved.

1. Introduction
Caesalpinia sappan L., called “Su Mu” in China, is a traditional
medicine distributed in China, India, Burma and Vietnam, whose
dried heartwood was traditionally used as a folk medicine for the
treatment of various diseases such as burning sensations, leprosy,
and dysentery. In addition, it has also been used as anti-inflammation
(Hikino et al., 1977; Hong et al., 2002; Bae et al., 2005; Shen et al.,
2007; Hu et al., 2009), antioxidant (Badami et al., 2003; Satri et al.,
n
Corresponding author. Tel.: +86 459 8153097; fax: +86 459 8153097.
E-mail address: sunshiqin7610@163.com (S.-Q. Sun).
1
These authors contributed equally to this work.
2003; Pan et al., 2004) and immunomodulation (Choi et al., 1997; Yu
et al., 2002; Zhou et al., 2003; Ye et al., 2006; Zheng et al., 2008), etc.
Washiyama et al. identified three compounds from Caesalpinia
sappan L., such as sappanchalcone, protosappanin D and protosap-
panin E, which suppress IL-6 mRNA expression (Washiyama et al.,
2009). In our previous study, we have demonstrated that the ethanol
extracts from dried heartwood of Caesalpinia sappan L. can signifi-
cantly attenuate collagen-induced arthritis (CIA) in rats by reducing
the levels of various inflammatory cytokines in serum such as IL-1β,
TNF-α and PEG2, especially the level of IL-6 was reduced drastically.
The level of IL-6 in the CIA rats treated with Caesalpinia sappan L.
extracts was significantly lower than that of normal rats (Wang et al.,
2011). It suggests that some compounds in the extracts may be
effective for the treatment of rheumatoid arthritis (RA).

0378-8741/$ - see front matter & 2013 Elsevier Ireland Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.jep.2013.03.050
38 M.-J. Chu et al. / Journal of Ethnopharmacology 148 (2013) 37–44
RA is a chronic inflammatory disease characterized by per-
sistent synovitis and destruction of bone and cartilage in multi-
ple joints. IL-6 is a pleiotropic cytokine with a wide range of
biological activities including a pro-inflammatory mediator
activity. It has been reported that the overproduction of IL-6
by activated synoviocytes, lymphocytes, fibroblasts, monocytes
and endothelial cells plays an important role in the prognosis of
RA (Fonseca et al., 2009). Interaction of IL-6 with the soluble IL-
6 receptor leads to osteoclast differentiation and may also lead
to joint destruction and osteoporosis associated with RA
(Nishimoto et al., 2007). Therefore, inhibition of IL-6 activity
may reduce the severity of the disease. In fact, biological agents
with IL-6 inhibitory activity such as tocilizumab, anti-IL-6
receptor monoclonal antibody and other biological agents have
been shown to improve systemic inflammatory symptoms in RA
(Klinkhoff, 2004; Nurmohamed, 2009; Yoshizaki et al., 1998).
However, it has also been suggested that these agents may cause
serious side effects including upper respiratory-tract infections,
headache, nasopharyngitis, and gastrointestinal events (Wen
et al., 2007). Thus successful identification and isolation of
active compounds from the herb to suppress the IL-6 production
might have a great therapeutic potential against RA. In fact,
crude extracts of Caesalpinia sappan L. reduced IL-6 production,
however, the main active compounds of Caesalpinia sappan L.
have not been identified. Thus the purpose of this work was to
identify the active compounds from Caesalpinia sappan L., which
inhibit IL-6 production.
The traditional and commonly used methods to identify
active compounds from the herb are costly and time-consum-
ing, which involve the separation of single compounds from
extraction, the evaluation of preliminary pharmacological
effect, and the screening of the active components. Recently,
partial least square (PLS) method was introduced as a new tool
for identification of active compounds from the herbal extracts.
PLS method is a supervised method which enables the com-
pression and visualization of the data in a reduced dimensional
space based on the dissimilarities among pharmacological
effects of extracts induced by different compounds
(Mcdonald et al., 2004). The fingerprint of the herbal extract
is established by UPLC–MS to present the characteristics of
individual compound. Moreover, fast qualitative and quantita-
tive analysis for the extract with complicated components can
be achieved by UPLC–MS (Liu et al., 2012). The inhibitory
activity of the extract on the IL-6 production can be evaluated
by using LPS-stimulated RAW cells (Liu et al., 2005).
Many methods to extract different proportions of compounds
from Caesalpinia sappan L. for analytical purposes have been
explored (Wu et al., 2011). The refluxing method has been used
to extract the anti-inflammatory active compounds from Caesalpi-
nia sappan L. (Wang et al., 2011). The supercritical CO2 fluid
extraction (SCFE) has also been widely applied in the herbal
extraction. Compared to traditional column chromatography
which is time-consuming, inefficient and requires a large volume
of toxic solvents, extraction with SCFE has many advantages such
as notoxicity, nonflammability, inability to leave residual chemical,
and moderate operating temperatures and pressures (Orio et al.,
2012). In our knowledge, extraction of compounds from Caesalpi-
nia sappan L. by SCFE has not been reported. In our present study,
we first prepared the extracts with different proportions of
compounds by SCFE and refluxing methods. Second, we estab-
lished UPLC–MS and in vitro pharmacological analysis of Caesalpinia
sappan L. extracts. Finally, PLS analysis was utilized as a data
reduction technique to generate a visual and objective plot to
discriminate active compounds based on these peak areas and
active information.
2. Materials and methods
2.1. Preparation of the herbal extracts
The heartwood of Caesalpinia sappan L. purchased from Beijing
Tong Ren Tang Group Co., Ltd., (Tong Ren Tang, China) was dried
and ground into a fine powder. Active components were extracted
by SCFE and refluxing methods. In SCFE (HA211-50-06, China)
method, 200 g dry powder from the heartwood of Caesalpinia
sappan L. was extracted at combinations of different temperatures
(30, 40, 50, 60 1C), pressures (15, 20, 25, 30 Mpa), and times (20,
40, 60 min). Ethanol (400 mL) was used as a co-solvent. The flow
rate was kept at 15 L/h by the CO2–ethanol. In refluxing method
(Wang et al., 2011), 5 g dried heartwood of Caesalpinia sappan L.
powder was extracted with different concentrations of 200 mL
ethanol as a solvent (0%, 30%, 60%, 90%) at different temperatures
(25, 45, 65, 85 1C) for a fixed time of 30 min. The total of 64
samples (48 from SCFE and 16 from refluxing) were evaporated
under a reduced pressure rotary evaporator to dryness and
collected. Each of 64 crude extracts was split into three parts for
the pharmacological activity analysis, for the UPLC–MS assay, and
finally for storage for reference purposes.
2.2. Pharmacological activity analysis of herbal extracts
2.2.1. Preparation of sample solutions
To prepare the sample solution for pharmacological activity
assay, 10 mg crude extract was freshly dissolved in 1.0 mL methanol.
The solution was mixed and filtered through a 0.22 mm syringe filter,
then diluted with DMEM (Solarbio, China) to a series of concentra-
tions (10, 15, 20, 25, 50 mg/mL). MTT (Solarbio, China) and LPS (E. coli
02B:B6) (Sigma, USA) were diluted with PBS (Hyclone, USA) and
were kept at −20 1C.
2.2.2. RAW 264.7 cells
RAW 264.7 murine macrophage cell line was obtained from the
China Cell Line Bank. The cells were grown in DMEM containing
10% FBS (Gibco, USA), 100 U/mL penicillin and 100 mg/mL strepto-
mycin. They were maintained at 37 1C in 5% CO2.
2.2.3. Cytotoxicity assay
Cytotoxicity studies induced by herbal extracts and LPS were
determined by MTT assay (Nandi et al., 2012). Briefly, cells were
seeded (2
105 cells/mL, 200 mL) in 96-well plates and incubated
overnight. Then the medium was replaced with diverse concentra-
tions of Caesalpinia sappan L. extracts (10, 15, 20, 25, 50 mg/mL), LPS
(10, 100, 1000 ng/mL), 0.2% methanol, or medium (n¼3). After 24 h,
20 μL MTT (5 mg/mL) was added to each well and the cells were
further incubated for 4 h at 37 1C in 5% CO2. MTT was removed and
cells were lysed with 150 μL/well of DMSO. The optical density (OD)
was measured at 492 nm on a microplate reader (Tecan, Austria).
2.2.4. Effects of herbal extracts on the IL-6 production
To determine the effects of herbal extracts on LPS-induced IL-6
production in RAW 264.7 cells, cells were stimulated with LPS (100 ng/
mL) in the presence or the absence of herbal extracts (20 mg/mL). In
brief, RAW cells (3
105 cells/mL in 1 mL medium) were seeded onto
24-well plate for 20 h at 37 1C, 5% CO2. Then medium was replaced
with various stimulants such as LPS with or without herbal extracts for
additional 20 h. Finally, cell-free medium was collected after a spin
(300 g, 5 min) and was kept at −20 1C. The IL-6 level in the collected
medium was determined by ELISA kit (RapidBio Lab, USA) according to
the manufacturer's protocol (Tao et al., 2009).
 M.-J. Chu et al. / Journal of Ethnopharmacology 148 (2013) 37–44 39

Table 1
Extraction conditions of Caesalpinia sappan L. by SCFE method.

Sample Extraction temperature Extraction pressure Extraction time Sample Extraction temperature Extraction pressure Extraction time
no. (1C) (Mpa) (min) no. (1C) (Mpa) (min)

1 30 15 20 25 50 15 20
2 30 15 40 26 50 15 40
3 30 15 60 27 50 15 60
4 30 20 20 28 50 20 20
5 30 20 40 29 50 20 40
6 30 20 60 30 50 20 60
7 30 25 20 31 50 25 20
8 30 25 40 32 50 25 40
9 30 25 60 33 50 25 60
10 30 30 20 34 50 30 20
11 30 30 40 35 50 30 40
12 30 30 60 36 50 30 60
13 40 15 20 37 60 15 20
14 40 15 40 38 60 15 40
15 40 15 60 39 60 15 60
16 40 20 20 40 60 20 20
17 40 20 40 41 60 20 40
18 40 20 60 42 60 20 60
19 40 25 20 43 60 25 20
20 40 25 40 44 60 25 40
21 40 25 60 45 60 25 60
22 40 30 20 46 60 30 20
23 40 30 40 47 60 30 40
24 40 30 60 48 60 30 60

2.3. LC–MS analysis of herbal extracts
2.3.1. Preparation of sample solutions
To prepare the herbal extracts for LC–MS analysis, 5 mg crude
extract was dissolved in 1.0 mL methanol. After filtration through
0.22 mm syringe filter, the sample was further diluted with methanol
to 1 mg/mL.
2.3.2. Optimization of LC–MS conditions
In order to establish more informative and reliable LC–MS total
ionic chromatographic fingerprints of Caesalpinia sappan L.
extracts, UPLC–MS fingerprint was applied to distinguish the
compounds of them. The chromatographic conditions such as
the choice of mobile phase, and the selection of injection quantity
were optimized based on analysis time, resolution, and the
number of peaks. Analysis was carried out on a Waters UPLC
systerm (AcquityTM, USA) equipped with sample manager, binary
solvent manager and column heater-cooler. The separation was
performed on a Waters C18 column (2.1
50 mm, 1.7 mm). The
mobile phase consisted of acetonitrile (A) and MilliQwater
(B) using a linear gradient program of 5–20% A in 0–10 min, 20–
30% A in 10–15 min, 30–70% A in 15–18 min, 70% A in 18–28 min,
70–100% A in 28–32 min. The column temperature and the flow
rate were set at 30 1C and 0.208 mL/min, respectively. Ten micro-
grams extract were injected to the column.
The above UPLC system was interfaced with a Waters TOF MS
equipped with an ESI source operating in negative ion detection mode
(Micromasss, USA). The acquisition parameters were drying gas, high
purity nitrogen (N2); drying gas temperature, 300 1C; drying gas flow-
rate, 750 L/h; source temperature, 110 1C; capillary voltage, 1800 V;
scan mass, 100–1000 m/z (Han et al., 2008). Data acquisition and
analysis were controlled by Masslynx Software (Version 4.1, USA).
2.3.3. UPLC–MS fingerprints of herbal extracts
The method validation of fingerprint analysis was performed
based on the RSD values of retention times and peak areas of all
the peaks from Caesalpinia sappan L. extract. We evaluated the
precision, reproducibility and stability of fingerprint method by
randomly selecting one extract (Liu et al., 2008) (SCFE, 30 1C,
20 Mpa, 20 min). The precision and reproducibility of method
were estimated by the analysis of six injections of sample solution
and six replicates of solid sample, respectively. We also assessed
the stability of method by the analysis of six injections of sample
solution in different time (0, 2, 4, 8, 12, 24 h).
In order to identify common peaks and characteristic peaks, the
fingerprints of 64 samples from different extract conditions were
analyzed. Among these fingerprints, some peaks which indicate a
high peak area and stable time in each fingerprint were selected as
the reference substances. Compared to the retention time and m/z,
peaks detected in all fingerprints of the 64 samples were assigned
as “common peaks” and peaks detected in limited number of
samples were assigned as “characteristic peaks”.
2.4. Analysis of active compounds from herbal extracts by PLS
analysis
Peak area information (X variables) and inhibitory rates
(Y variable) of 64 samples from different extract conditions were
applied to the PLS analysis to establish the multiple linear
regression models. The inhibitory rate (% IR) of 64 samples was
calculated as follows:
%IR ¼ 100
ðAcontrol −Asample Þ=Acontrol
where Acontrol and Asample are the IL-6 concentrations of the control
(cultures stimulated with LPS) and the sample (cultures stimulated
with LPS in the presence of extracts), respectively.
In PLS modeling, score plot of PLS was employed to remove
specific variables. Then PLS modeling was carried out again with
the remaining variables. In addition, an observed versus predicted
values plot was produced to assess the prediction capabilities. The
information including regression coefficients and loading plot
have been used to select original regression variables to eliminate
some unimportant or uninformative variables and to obtain the
simpler model without losing prediction power (Chen et al., 2009).
We have selected the top five chromatographic peaks most
responsible for the inhibitory activity based on regression coeffi-
cients. Based on the LC–MS analysis and comparison with
40 M.-J. Chu et al. / Journal of Ethnopharmacology 148 (2013) 37–44

Table 2
Extraction conditions of Caesalpinia sappan L. by refluxing method.

Sample no. Alcohol concentration (%) Temperature (1C) Time (min) Sample no. Alcohol concentration (%) Temperature (1C) Time (min)

49 0 25 30 57 60 25 30
50 0 45 30 58 60 45 30
51 0 65 30 59 60 65 30
52 0 85 30 60 60 85 30
53 30 25 30 61 90 25 30
54 30 45 30 62 90 45 30
55 30 65 30 63 90 65 30
56 30 85 30 64 90 85 30

Fig. 1. Inhibition of various Caesalpinia sappan L. extracts (1–64) on IL-6 production in LPS-induced RAW 264.7 cells. RAW 264.7 cells (3
105) were treated with 20 mg/mL
Caesalpinia sappan L. extracts in the presence of 100 ng/mL LPS for 20 h. The supernatants were collected and assayed for IL-6 level by ELISA. Results were analyzed as IL-6
level by %-LPS-treated control. Mean and SD values from n¼ 4 are shown. nstatistical significance (p o 0.05).

Fig. 2. The variations of compounds among 64 samples produced in different extract conditions by UPLC–MS. The separation was performed on a Waters C18 column
(2.1
50 mm, 1.7 mm). The mobile phase consisted of acetonitrile (A) and MilliQwater (B) using a linear gradient program of 5–20% A in 0–10 min, 20–30% A in 10–15 min,
30–70% A in 15–18 min, 70% A in 18–28 min, 70–100% A in 28–32 min. The column temperature and the flow rate were set at 30 1C and 0.208 mL/min, respectively.
Ten micrograms extract was injected into the column.

literature data (Nguyen et al., 2005; Zhao et al., 2010), the main programs were operated by SIMCA-P (Version 12.0, Demo, Ume-
compounds of the chromatographic peaks were speculated. All trics, Umea, Sweden) software.
 M.-J. Chu et al. / Journal of Ethnopharmacology 148 (2013) 37–44 41

2.5. Statistical analysis suppression of LPS-induced macrophage inflammatory response,

All values were expressed as mean 7SD. Data were statistically
analyzed by student t-test using SPSS 16.0 software. In all com-
parisons, statistical difference was accepted at p o0.05.
3. Results
3.1. Caesalpinia sappan L. extracts had no cytotoxicity but inhibited
LPS-induced IL-6 production
First we described the SCFE and refluxing methods to obtain
Caesalpinia sappan L. extracts (1–64) in Tables 1 and 2.
We then evaluated whether these extracts (1–64) have any effects
on LPS-induced IL-6 production in RAW 264.7 cells. RAW cells were
stimulated with LPS in the presence or the absence of each herbal
extract (1–64). IL-6 production was determined after 20 h incubation
as described in Materials and methods. As shown in Fig. 1, although
the magnitude may vary, all the extracts significantly reduced LPS-
induced IL-6 production in RAW cells. We presented IL-6 production
by control (LPS alone) as 100%. Thus Herbal extracts induced a
which was not associated with its cytotoxicity (data not shown).
3.2. LC–MS fingerprints of Caesalpinia sappan L. extracts
As shown in Fig. 2, there were 457 peaks detected in 64
fingerprints of Caesalpinia sappan L. extracts, more than 50 peaks
were separated in every fingerprint by UPLC–MS analysis. The
precision of the proposed method, based on analysis of six replicate
samples, was below 0.4% and 1.5% for RSD of retention times and
peak areas of all peaks, respectively. The reproducibility of the
method was determined by six sample solutions from one randomly
selected extract. The RSD of the retention times and the peak areas
were both less than 3%. The stability test was performed with sample
solutions for 24 h. RSDs of the retention times and the peak areas
were less than 3%. Total ionic chromatographic fingerprints of
the extracts showed good precision, reproducibility and stability.
All results indicated that the conditions for the fingerprint analysis
were adequate, valid and applicable (data not shown).
The fingerprints of 64 samples after UPLC–MS are shown in
Fig. 2. The peak, which was present in all 64 samples with the
highest intensity, was selected as the internal reference substance.

Fig. 3. Observed versus predicted inhibitory activity of Caesalpinia sappan L. extracts on IL-6 production suggests good predictive performance of the PLS model. R2 ¼ 0.9041.

Fig. 4. Loading plot shows the correlation of the inhibitory activity (Y variable) and chemical component (X variable). Proximity to the inhibitory activity reflects the strength
of association of individual chemical components to the activity.
42 M.-J. Chu et al. / Journal of Ethnopharmacology 148 (2013) 37–44
The retention time of the peak was 23.3 70.08 min. Then the
specific peaks from different fingerprints that have the same m/z
and retention times varying within the SD value of the reference
substance's retention time were recognized as the same com-
pounds (Leston et al., 2011). Finally, 457 peaks were distinguished
from 64 fingerprints and were marked as variables including
common peaks and characteristic peaks.
3.3. Active compounds analysis of herbal extracts
To confirm the inhibitory compounds of Caesalpinia sappan L.
on IL-6 production, the PLS analysis was initially performed with
all 457 compositional variables versus the inhibitory rate variable.
Then we removed the samples with an extreme position in the
score plot. The PLS modeling was established until all samples
clustered within the ellipse based on the score plot (Supplemen-
tary data 1.a-d). The number of compositional variables was
reduced from 457 to 435, and PLS analysis was carried out with
these remaining 435 variables. As shown in Fig. 3, the R2 value
(0.9041) strongly suggests that our PLS analysis is reliable and that
we can likely perform this analysis to predict active compounds
without additional time-consuming and costly experiments.
Once the predictive model was determined, the strength of
association (PLS loadings) between “the chemical components” and
“the inhibitory activity” was evaluated in a loading plot (Fig. 4).This
plot shows the clustering of inhibitory activity and chemical compo-
nent variables, illustrating the chemical components (X variables)
most closely associated with inhibitory activity (Y variable). We can
roughly analyze the active compound with stronger IL-6 inhibitory
activity based on the distance between X variable and Y variable.
Additionally, based on list of coefficients from PLS analyses (Supple-
mentary data 2), the contributions of the peaks were evaluated by
the final PLS model. The components with higher regression coeffi-
cients were predicted as active compounds. These compounds with
Fig. 5. The peaks such as 71, 93, 150, 157 and 168 identified by PLS analysis were
shown in the fingerprint (6–16 min) of Caesalpinia sappan L. extract obtained by
UPLC–MS.
Table 3
Main active compounds identified from the mass analysis.
Peak no. tR (min) Regression coefficient
71 6.5 70.0279 0.0346
93 8.9 70.0434 0.0352
150 11.5 70.0170 0.0420
157 12.7 70.0213 0.0307
168 13.0 70.0130 0.0359
the stronger IL-6 inhibitory activity were the peaks 71, 93, 150, 157
and 168 (Fig. 4). Other components had little or negative correlation
with the inhibitory activity.
The more accurate molecular mass of these five peaks were
obtained by ESI-TOF-MS method. The mass data (m/z) [M−H]- for
the peaks “71”, “93”, “150”, “157” and “168” were “283”, “585”,
“569”, “581” and “553”, respectively. These data strongly suggest
the similarities of our peaks “71”, “93” and “150” to “sappanone A”,
“protosappanin E” and “neoprotosappanin”, respectively (Nguyen
et al., 2005; Zhao et al., 2010). However, no similarities were found
for our peaks “168” and “283”. The fingerprint analysis shows the
chromatographic features of those five peaks (Fig. 5). The mass
spectrum information about these peaks can be found in Table 3.
4. Discussion
The heartwood of Caesalpinia sappan L. is a common remedy in
traditional medicine and possesses diverse biological activities
including anti-inflammatory properties (Wu et al., 2011). It is widely
recognized that IL-6 is involved in inflammation thus it can be a
therapeutic target. We performed an in vitro assay for ethanolic
extracts of Caesalpinia sappan L. that inhibit the IL-6 production. As
shown in Fig. 1, our results revealed that all ethanolic extracts of
Caesalpinia sappan L. had inhibitory activities on the LPS-induced IL-6
production in RAW 264.7 macrophages, but their effectiveness
differed for the different extraction conditions. We speculated there
are different inhibitory constituents from Caesalpinia sappan L.
extracts on IL-6 production. This finding might contribute to provide
the basis for the optimization of extraction conditions of Caesalpinia
sappan L. The results from Fig. 1 showed that the extract 14, 15, 17, 21,
32, 39, 45, 47 and 62 had the relatively higher inhibitory effect on
IL-6 production. Eight extracts except extract #62 were prepared by
SCFE method rather than refluxing method (Tables 1, 2). We
conclude the SCFE method (extract 1-48) has more advantage over
the refluxing method (extract 49-64) for extracting active com-
pounds of Caesalpinia sappan L. suppressing IL-6 production. Addi-
tionally, we found the extract 15 contained more kinds of identified
active compounds than the remaining seven extracts by SCFE
method. So the optimal extraction condition may be at 40 1C,
15 Mpa, and 60 min of extraction time with 100% ethanol (extract
#15). We believe it is very important to adjust procedures to
maximize the extraction of useful compounds from Chinese medic-
inal materials minimizing any degradation or even loss.
In traditional method, we identify the active compounds on the
basis of pharmacological effects of all the isolated monomers, which
is blind, costly and time-consuming. Thus we utilized simplified
method to analyze the extracts of Caesalpinia sappan L. to find active
compounds. Partial least-squares discriminant analysis (PLS-DA)
has been commonly utilized to establish a classification model for
the determination of active compounds (Chen et al., 2008). PLS-DA
requires multiple groups (i.e. higher inhibitory group (430%
inhibition) and lower inhibitory group (o30% inhibition)). How-
ever, our 64 extracts only had higher inhibitory rates and the PLS-
Main mass data (m/z) Identification
283 [M−H]− Sappanone A
585 [M−H]−, 283 Protosappanin E
569 [M−H]−, 375 Neoprotosappanin
581 [M−H]−, 283 Unknown
553 [M−H]− Unknown
 M.-J. Chu et al. / Journal of Ethnopharmacology 148 (2013) 37–44
DA analysis was not applicable. Here, we applied the PLS method to
dea1 with data with many variables such as a large number of
compounds from various extracts. By comparing the fingerprints of
64 different extracts, the 457 peaks detected in all fingerprints
(Fig. 2) were recognized as different variables. The regression
coefficients obtained by the PLS analysis revealed five active
compounds that are likely responsible for an inhibitory activity on
IL-6 production in LPS-stimulated RAW 264.7 cells. As shown in
Table 3, the exact molecular mass of the active components was
obtained by ESI-TOF-MS experiment. The three compounds such as
sappanone A, protosappanin E and neoprotosappanin were identi-
fied. We are the first group to report the identification of the main
active compounds with inhibitory effects on IL-6 production from
Caesalpinia sappan L. based on PLS method. Our finding may lead to
a discovery of new active compounds from the herb to treat RA.
In traditional method of identifying active compounds of Caesalpinia
sappan L., Washiyama et al. (2009) performed an in vitro assay to identify
compounds from methanolic extract of Caesalpinia sappan L. that inhibit
the inflammatory mediators using the J774.1 cell line. They identified
three compounds such as sappanchalcone, protosappanin D and proto-
sappanin E, which reduced mRNA expression of IL-6, TNF-α, COX-2 and
iNOS. These compounds also inhibited both NO and PGE2 production in
J774.1 cells. Although other compounds are present in Caesalpinia
sappan L., there are very limited numbers of reports on compounds
with anti-inflammatory activities to suppress IL-6 production. It is
possible that active components may be degraded or lost due to
complication of the traditional extraction methods. We identified
sappanone A, protosappanin E and neoprotosappanin to inhibit IL-6
production. Thus there is discrepancy between our findings and
Washiyama's findings. First of all, sappanone A and neoprotosappanin
were not identified by Washiyama et al. or any other group. We
hypothesize that our gentle and simple procedures enabled us to
identify these compounds. Although we identified protosappanin E to
inhibit IL-6 production, Washiyama et al. only identified its inhibitory
activity of IL-6 mRNA expression in J774.1 cells. We are unable to
speculate whether these differences are due to cell line or technical
issues. However, these data agree to the point that protosappanin E does
inhibit IL-6 either at RNA level or protein level. Again, our simplified and
cost-effective methods have advantages over traditional methods to
identify compounds that could be overlooked or degraded during harsh
and time-consuming traditional methods. Further studies are needed to
confirm possible mechanism.
5. Conclusions
Our study strongly suggests that neoprotosappanin, protosap-
panin E, sappanone A and two other unidentified compounds are
the main gradients that could inhibit IL-6 production and may lead
to a new treatment for rheumatoid arthritis. We are in the process
of determination of the other two unidentified. Similarly, we plan
to determine the effectiveness of these five compounds on IL-6
production (in vitro) and on the treatment for rheumatoid arthritis
(in vivo). Finally, we hope our simplified method will benefit the
identification of the active compounds of traditional Chinese
medicine and further instruct the isolation of active compounds
to lead new treatments for various diseases.
Acknowledgments
This work was supported by Heilongjiang Postdoctoral Financial
Assistance (No. LBH-Z06245) and the Innovative Fund of Harbin
Medical University Graduate Student (HCXS20100019).
43
Appendix A. Supporting information
Supplementary data associated with this article can be found in
the online version at http://dx.doi.org/10.1016/j.jep.2013.03.050.
References
Badami, S., Moorkoth, S., Rai, S.R., Kannan, E., Bhojraj, S., 2003. Antioxidant activity of
Caesalpinia sappan heartwood. Biological & Pharmaceutical Bulletin 26, 1534–1537.
Bae, I.K., Min, H.Y., Han, A.R., Seo, E.K., Lee, S.K., 2005. Suppression of
lipopolysaccharide-induced expression of inducible nitric oxide synthase by
brazilin in RAW 264.7 macrophage cells. European Journal of Pharmacology
513, 237–242.
Chen, K.W., Reiman, E.M., Huan, Z.D., Caselli, R.J., Bandy, D., Ayutyanont, N.,
Alexander, G.E., 2009. Linking functional and structural brain images with
multivariate network analyses: a novel application of the partial least square
method. Neuroimage 47, 602–610.
Chen, Y., Zhu, S.B., Xie, M.Y., Nie, S.P., Liu, W., Li, C., Gong, X.F., Wang, Y.X., 2008.
Quality control and original discrimination of Ganoderma lucidum based on
high-performance liquid chromatographic fingerprints and combined chemo-
metrics methods. Analytica Chimica Acta 623, 146–156.
Choi, S.Y., Yang, K.M., Jeon, S.D., Kim, J.H., Khil, L.Y., Chang, T.S., Moon, C.K., 1997.
Brazilin modulates immune function mainly by augmenting T cell activity in
halothane administered mice. Planta Medica 63, 405–408.
Fonseca, J.E., Santos, M.J., Canhão, H., Choy, E., 2009. Interleukin-6 as a key player in
systemic inflammation and joint destruction. Autoimmunity Reviews 8, 538–542.
Han, C., Shen, Y., Chen, J.H., Chun Lee, F.S., Wang, X.R., 2008. HPLC fingerprinting
and LC–TOF-MS analysis of the extract of Pseudostellaria heterophylla (Miq.)
Pax root. Journal of Chromatography B 862, 125–131.
Hikino, H., Taguchi, T., Fujimura, H., Hiramatsu, Y., 1977. Antiinflammatory princi-
ples of Caesalpinia sappan wood and of Haematoxylon campechianum wood.
Planta Medica 31, 214–220.
Hong, C.H., Hur, S.K., Oh, O.-J., Kim, S.S., Nam, K.A., Lee, S.K., 2002. Evaluation of
natural products on inhibition of inducible cyclooxygenase (COX-2) and nitric
oxide synthase (iNOS) in cultured mouse macrophage cells. Journal of Ethno-
pharmacology 83, 153–159.
Hu, C.M., Liu, Y.H., Cheah, K.P., Li, J.S., Lam, C.S.K., Yu, W.Y., Choy, C.S., 2009. Heme
oxygenase-1 mediates the inhibitory actions of brazilin in RAW 264.7 macrophages
stimulated with lipopolysaccharide. Journal of Ethnopharmacology 121, 79–85.
Klinkhoff, A., 2004. Biological agents for rheumatoid arthritis. Drugs 64, 1267–1283.
Leston, S., Nunes, M., Freitas, A., Barbosa, J., Ramos, F., Pardal, M.A., 2011. A LC–MS/
MS methodology to determine furaltadone residues in the macroalgae Ulva
lactuca. Journal of Chromatography B 879, 3832–3836.
Liu, F., Ren, D.Q., Guo, D.A., Pan, Y.F., Zhang, H.Z., Hu, P., 2008. Method development
for gypenosides fingerprint by high performance liquid chromatography with
diode-array detection and the addition of internal standard. Chemical &
Pharmaceutical Bulletin 56, 389–393.
Liu, J.J., Li, X., Yue, Y., Li, J., He, T., He, Y.Z., 2005. The inhibitory effect of quercetin on
IL-6 production by LPS stimulated neutrophils. Cellular & Molecular Immunology
2, 455–459.
Liu, X.Y., Zhang, S.S., Lu, X.M., Zheng, S.N., Li, F.M., Xiong, Z.L., 2012. Metabonomic
study on the anti-osteoporosis effect of Rhizoma drynariae and its action
mechanism using ultra-performance liquid chromatography-tandem mass
spectrometry. Journal of Ethnopharmacology 139, 311–317.
Mcdonald, J.D., Eide, I., Seagrave, J., Zielinska, B., Whitney, K., Lawson, D.R.,
Mauderly, J.L., 2004. Relationship between composition and toxicity of motor
vehicle emission samples. Environmental Health Perspectives 112, 1527–1538.
Nandi, D., Besra, S.E., Vedasiromoni, J.R., Giri, V.S., Rana, P., Jaisankar, P., 2012. Anti-
leukemic activity of Wattakaka volubilis leaf extract against human myeloid
leukemia cell lines. Journal of Ethnopharmacology 144, 466–473.
Nguyen, M.T.T., Awale, S., Tezuka, Y., Tran, Q.L., Kadota, S., 2005. Xanthine oxidase
inhibitors from the heartwood of Vietnamese Caesalpinia sappan. Chemical &
Pharmaceutical Bulletin 53, 984–988.
Nishimoto, N., Hashimoto, J., Miyasaka, N., Yamamoto, K., Kawai, S., Takeuchi, T.,
Murata, N., Heijde, D., Kishimoto, T., 2007. Study of active controlled mono-
therapy used for rheumatoid arthritis, an IL-6 inhibitor (SAMURAI): evidence of
clinical and radiographic benefit from an x ray reader-blinded randomised
controlled trial of tocilizumab. Annals of the Rheumatic Diseases 66, 1162–1167.
Nurmohamed, M.T., 2009. Newer biological agents in the treatment of rheumatoid
arthritis. Drugs 69, 2035–2043.
Orio, L., Alexandru, L., Cravotto, G., Mantegna, S., Barge, A., 2012. UAE, MAE, SFE-CO2
and classical methods for the extraction of Mitragyna speciosa leaves. Ultra-
sonics Sonochemistry 19, 591–595.
Pan, Y.M., Liang, Y., Wang, H.S., Liang, M., 2004. Antioxidant activities of several
Chinese medicine herbs. Food Chemistry 88, 347–350.
Satri, R., Tarigan, P., Freisleben, H.J., Rumampuk, R.J., Murakami, A., 2003. Antiox-
idant activity in vitro of two aromatic compounds from Caesalpinia sappan L.
BioFactors 19, 71–77.
Shen, J., Zhang, H.Y., Lin, H., Su, H., Xing, D.M., Du, L.J., 2007. Brazilein protects the
brain against focal cerebral ischemia reperfusion injury correlating to inflam-
matory response suppression. European Journal of Pharmacology 558, 88–95.
44 M.-J. Chu et al. / Journal of Ethnopharmacology 148 (2013) 37–44
Tao, J.Y., Zheng, G.H., Lei Zhaoc, Wu, J.G., Zhang, X.Y., Zhang, S.L., Huang, Z.J., Xiong,
F.L., Li, C.M., 2009. Anti-inflammatory effects of ethyl acetate fraction from
Melilotus suaveolens Ledeb on LPS-stimulated RAW 264.7 cells. Journal of
Ethnopharmacology 123, 97–105.
Wang, Y.Z., Sun, S.Q., Zhou, Y.B., 2011. Extract of the dried heartwood of Caesalpinia
sappan L. attenuates collagen-induced arthritis. Journal of Ethnopharmacology
136, 271–278.
Washiyama, M., Sasaki, Y., Hosokawa, T., Nagumo, S., 2009. Anti-inflammatory
constituents of Sappan lignum. Biological & Pharmaceutical Bulletin 32, 941–944.
Wen, C.P., Jin, C.Y., Xu, Z.L., Cao, L.Y., 2007. Clinical observation of Jiedu Tongluo Lishi
decoction on treating reactive rheumatoid arthritis. Zhongguo Zhong Yao Za Zhi
32, 1306–1310.
Wu, S.Q., Otero, M., Unger, F.M., Goldring, M.B., Phrutivorapongkul, A., Chiari, C.,
Kolb, A., Viernstein, H., Toegel, S., 2011. Anti-inflammatory activity of an
ethanolic Caesalpiniasappan extract in human chondrocytes and macrophages.
Journal of Ethnopharmacology 138, 364–372.
Ye, M., Xie, W.D., Lei, F., Meng, Z., Zhao, Y.N., Su, H., Du, L.J., 2006. Brazilein, an
important immunosuppressive component from Caesalpinia sappan L. Interna-
tional Immunopharmacology 6, 426–432.
Yoshizaki, K., Nishimoto, N., Mihara, M., Kishimoto, T., 1998. By blocking IL-6 signal
transduction with a humanized anti-IL-6 receptor antibody. Springer Seminars
in Immunopathology 20, 247–258.
Yu, B., Hou, J.B., Liu, H., Xu, W., Cui, L.L., 2002. Effects of different competents of
Caesalpinia sappan L. on immunocompetence of rat immunocytes. Chinese
Journal of Critical Care Medicine 22, 187–189.
Zhao, H.X., Bai, H., Li, W., Li, J., Wang, Y.S., 2010. Study on Chemical Constituents of
Caesalpinia sappan L. Food and Drug 12, 176–179.
Zheng, J.X., Zhou, Y.B., Liu, Y.Z., Ke, R., Sun, J., 2008. Effect of ethyl acetate extract of
sappan wood on expression of myocardial GrB mRNA in rat model of allogeneic
ectopic cardiac transplantation. Chinese Journal of Integrated Traditional and
Western Medicine 28, 537–540.
Zhou, Y.B., Yao, F.Z., Han, J.R., Yang, Y., Zhang, C.F., Shi, Y.P., 2003. Effect of sappan
wood on perforin mRNA expression in myocardium of rats after allogeneic
cardiac transplantation. Chinese Journal of Integrated Traditional and Western
Medicine 23, 370–372.