Journal of Pharmaceutical and Biomedical Analysis 96 (2014) 104–110
Contents lists available at ScienceDirect
Journal of Pharmaceutical and Biomedical Analysis journal homepage: www.elsevier.com/locate/jpba
Two-dimensional correlation infrared spectroscopy applied to analyzing and identifying the extracts of Baeckea frutescens medicinal materials Adiana Mohamed Adib ∗ , Fadzureena Jamaludin, Ling Sui Kiong, Nuziah Hashim, Zunoliza Abdullah Natural Products Division, Forest Research Institute Malaysia, 52109 Kepong, Selangor, Malaysia
a r t i c l e
i n f o
Article history: Received 25 September 2013 Received in revised form 13 February 2014 Accepted 15 March 2014 Available online 25 March 2014 Keywords: FT-IR Two-dimensional correlation infrared spectroscopy Second derivative IR spectra Baeckea frutescens
a b s t r a c t Baeckea frutescens or locally known as Cucur atap is used as antibacterial, antidysentery, antipyretic and diuretic agent. In Malaysia and Indonesia, they are used as an ingredient of the traditional medicine given to mothers during confinement. A three-steps infra-red (IR) macro-fingerprinting method combining conventional IR spectra, and the secondary derivative spectra with two dimensional infrared correlation spectroscopy (2D-IR) have been proved to be effective methods to examine a complicated mixture such as herbal medicines. This study investigated the feasibility of employing multi-steps IR spectroscopy in order to study the main constituents of B. frutescens and its different extracts (extracted by chloroform, ethyl acetate, methanol and aqueous in turn). The findings indicated that FT-IR and 2D-IR can provide many holistic variation rules of chemical constituents. The structural information of the samples indicated that B. frutescens and its extracts contain a large amount of flavonoids, since some characteristic absorption peaks of flavonoids, such as ∼1600 cm−1 , ∼1500 cm−1 , ∼1450 cm−1 , and ∼1270 cm−1 can be observed. The macroscopical fingerprint characters of FT-IR and 2D-IR spectra can not only provide the information of main chemical constituents in medicinal materials and their different extracts, but also compare the components differences among the similar samples. In conclusion, the multi-steps IR macro-fingerprint method is rapid, effective, visual and accurate for pharmaceutical research. © 2014 Elsevier B.V. All rights reserved.
1. Introduction Baeckea frutescens or locally known as Cucur atap is a small tree which is found in Peninsular Malaysia and Sumatra. It is also found distributed along the coastal areas of Southern China and Australia. It grows wild on arid soils in the mountain and midlands. In Peninsular Malaysia, it is found both on the mountain and sandy coasts [1]. The tree trunk is short with several upright branches that spread out and droop at the ends as fine twigs. The leaves are opposite, small and very narrow, needle-like, only about 6–15 mm long. In traditional usage, it is claimed to be effective in treating influenza, dyspepsia, jaundice, dysentery, measles and irregular menstrual cycles. Its external usage is in treating furunculosis and impetigo [1]. They are also used in massaging postpartum women for the treatment of body aches and numbness of the
∗ Corresponding author. Tel.: +60 3 62797366/+60 137834263; fax: +60 3 62729805. E-mail addresses: [email protected], [email protected] (A.M. Adib). http://dx.doi.org/10.1016/j.jpba.2014.03.022 0731-7085/© 2014 Elsevier B.V. All rights reserved.
limbs [1]. B. frutescens leaf extract has been reported to have various pharmacological activity such as cytotoxic [2], anticariogenic [3] and antibabesial activities [4]. Chemical studies on the leaves of B. frutescens have indicated the presence of volatile oil [5], sesquiterpenes [6], chromone C-glycosides [7], phloroglucinols [2] and flavanones [8]. Currently, the main method for identifying the qualities of medicinal materials and extract products is to use chromatography examining the content of certain chemical constituent in the tested sample. It is well known that medicinal materials comprise hundreds of components, and produce their curative effects through synergistic effects of many ingredients, so the limited numbers of specific components cannot availably reflect the real qualities of the herbal medicines. Therefore, we need to find a quick and effective analysis to entirely monitor and reflect the whole constituents of the medicinal materials and their corresponding extract products. In recent years, there are many reports on the application of FT-IR in medicinal plants research on various aspects such as identification, quality control and forecasting stability [9–11]. For a complex system, FT-IR has shown advantages over the other
A.M. Adib et al. / Journal of Pharmaceutical and Biomedical Analysis 96 (2014) 104–110
105
1619
3374
1036 1449 1376 1236 1517 1161 1542 1319 1106
2926 2870
616 764
(a)
2928
819 1689
3388
2871
1238 1616
3352
1377 1138 1456 1274 1167 1386 1036 1189 1104 1316 1515
1613 1657
Absorbance (A)
997 975 889
(b)
2978
724 662 608 828 766 570
464
1202 1449 1350
1235
1501 1536
1704
1167
1038 1086 916 966
3398
667
814
717 643 764 678 576
(c) 1614 2929
1657
1450
1204 1380
1235 1142
1690
1039
1515 817
3391
765
612
(d)
1614 2936
1449
1374
1236
1070 1045
1515
915
(e)
4000.0
3600
3200
2800
2400
2000
1800
1600
1400
1200
1000
768 818
800
716
614
600
450.0
Wavenumber (cm-1)
Fig. 1. FT-IR spectra of B. frutescens and its different extracts: (a) B. frutescens; (b) dichloromethane extract; (c) ethyl acetate extract; (d) methanol extract; (e) aqueous extract.
conventional analysis methods, as it presents the whole features as well as the macro-fingerprint characters of the sample. Owing to the fingerprint characters and extensive applicability to the samples, FT-IR has played an important role in pharmaceutical analysis in recent years. Two-dimensional correlation infrared (2DIR) spectroscopy was proposed by Noda [12]. The construction of 2D correlation IR spectra is based on the detection of dynamic changes of a system under an external perturbation. Such a perturbation induces selective changes in molecular constituents associated with individual normal modes of vibration in the system. The correlation analysis is applied to a set of spectra taken during the perturbation, so as to yield 2D correlation IR spectra. There has been a rapid progress in the technique of 2D correlation spectral analysis. The external perturbations used to produce 2D correlation IR spectra were sinusoids in early days [13]. Nowadays, however, arbitrary wave forms with non-periodical perturbations are commonly used [14]. The applied perturbations can be light, heat, electricity, magnetism, chemistry, concentration changes, or mechanical force [15–17]. As long as the samples are different in components or contents, the differences will embody in their FT-IR spectra. In this paper, we report the main components and the holistic variation rules of chemical constituents in B. frutescens and its extracts by applying FT-IR, second derivative infrared and
2D-IR spectroscopy. The purpose of this study is to develop a rapid, accurate and feasible analytical method to appraise integrally the inherent qualities of herbal material and its corresponding extracts. 2. Experimental 2.1. Apparatus IR spectra were recorded on a Spectrum 100 Fourier transform infrared (FTIR) spectrometer (Perkin-Elmer, Santa Clara, CA, USA), equipped with a mid-infrared deuterated triglycine sulfate (DTGS) detector. The spectra were obtained from the scan range of 4000 to 450 cm−1 with a resolution of 4 cm−1 and with a total accumulation of 16 scans. Portable programmable temperature controller (4000 series TM High Stability Temperature Controller, Specac, Ltd.) was used in the range of 50–120 ◦ C. 2.2. Plant material Samples of B. frutescens were collected in January 2010 from Forest Research Institute Malaysia (FRIM) Research Station in Setiu, Terengganu, Malaysia. The plant was identified at Biological Resources Programme FRIM and the voucher specimen was
Table 1 The tentative assignment of Baeckea frutescen IR spectrum [18]. Band (cm−1 )
Base group and vibration mode
Main attribution
3374 2926 2870 1619 1517 1449 1319 1236 1106 1036
(O–H) as (C–H) s (C–H) (C=O), (C=C) rf (ar), rf (ha) rf (ar), rf (ha) I,as (=C–O–C) ıs (C–CO), (C–O), II,as (=C–O–C), f (C–C) ı(C–OH), f (C–C) ı(C–OH), I (=C–O–C), f (C–C)
OH CH3 , CH2 CH3 , CH2 Flavonoids, chromones, phlororoglucinols Flavonoids, chromones, phlororoglucinols Flavonoids, chromones, phlororoglucinols Flavonoids, chromones, phlororoglucinols Flavanoids, chromones phlororoglucinols, branched chain hydrocarbon Compound which has secondary alcohol functional group Flavonoids, compound which has primary alcohol functional group
Note: , stretching or frame vibration; ı, bending; s, symmetrical; as, asymmetrical; ar, aromatic, ha, heteroaromatic; rf, ring frame; f, frame; I, II, stretching mode.
106
A.M. Adib et al. / Journal of Pharmaceutical and Biomedical Analysis 96 (2014) 104–110
measurement of 2D-FTIR spectra, each sample disc was put into the sample pool connected with a temperature controller. Dynamic spectra were collected at different temperatures ranging from 50 to 120 ◦ C at an interval of 10 ◦ C. 2D-IR correlation spectra were acquired by treatment of the series of temperature-dependant dynamic spectra with 2D-IR correlation analysis employing Softdoc software developed by Tsinghua University (Beijing, China).
OH OH HO
Glc HO
O
O
OH O Rha
OH O
OH O (1)
(2)
3. Results and discussion
Fig. 2. Myricetin 3-O-␣-L-rhamnopyranoside (1) and 8--C-glucopyranosyl-5,7dihydroxy-2-isopropylchromone (2).
3.1. 1D-FTIR spectral analysis of B. frutescens leaves Fig. 1 shows the FT-IR spectra of B. frutescens leaves and its different extract at room temperature. The leaves of B. frutescen and its different extracts are complex mixtures which showed IR spectra with overlapping absorption peaks. In the spectra, we can see clearly that although they show substantial overlap of each absorption spectrum of various components, each band represents an overall overlap of some characteristic absorption peaks of functional groups in the samples. The FT-IR spectra of ethyl acetate extract (c) and methanol extract (d) are rather similar, but obviously differ from the spectra of B. frutescens leaves (a), chloroform extract (b) and aqueous extract (e). Considering B. frutescens leaves (Fig. 1a) is a mixture of complex compounds, its infrared spectrum is complex and contains several bands from the contribution of different functional groups. As expected the FT-IR spectrum shows a total overlap of each absorption spectrum of various components. The chemical compounds of B. frutescens leaves are sesquiterpenes [6], chromone C-glycosides [7], phloroglucinols [2] and flavonoids [8]. The principal components in B. frutescens leaves all have their own infrared characteristic peaks. As a kind of macro-fingerprinting features of natural product complex, several characters can be extracted such as the strongest peak at 3374 cm−1 belonging to the stretching vibration of O–H groups, the peak 2926 cm−1 and 2870 cm−1
deposited at the Natural Products Division Reference Collection, FRIM Kepong. The leaves of the plants were removed and used for macro-fingerprinting analysis. 2.3. Procedure The dried leaves of B. frutescens (2 kg) were pulverized into coarse powder. One-tenth of the coarse powder was further ground into fine powder form (sample a). The remaining coarse powder was orderly extracted by chloroform, ethyl acetate, methanol and water for three times, respectively. All the extracts were concentrated under vacuum and dried to yield the following samples; (b) chloroform (19.2 g); (c) ethyl acetate (27.4 g) and (d) methanol (35.7 g) and aqueous (31.2 g) extracts. After that, exactly 2 mg of each sample was mixed with 100 mg of potassium bromide (KBr) and the mixture was further ground and pressed under 10 t of pressure to produce a thin disc with 13 mm of diameter. 1D-FTIR spectra were recorded from a total of 16 scans in the 4000 to 450 cm−1 range with a resolution of 4 cm−1 . The second derivative spectra were obtained by using Savitzky–Golay filter through 13 point smoothing. Savitzky–Golay smoothing aimed for minimum distortion by least squares fitting a cubic polynomial. For the 1673 1639
1176 1558
1459
1720
1542
1628 1619
1698
1013
1484
1586
1735 1753
1787
1050 1152
1523
1434 1418
12781260
1409
1367
1215
1065
1333
1506
11241105
1299
1078
1029
1001
1710
Absorbance (A)
1449
1677
1642 1153
1478 1753
1786
1765
1734
1523 1576 15631553
1704
1538
1420 1435
1505
1776
1225
1394 1372 1336 1360
1177
12651246 1279
1051 1066
1013
1110 1137
1459
1030
1616
1800.0
1750
1700
1650
1600
1550
1500
1450
1400
1350
1300
1250
1200
1150
Wavenumber (cm-1)
Fig. 3. Second derivative spectra of the ethyl acetate extract and methanol extract.
1100
1050
1000.0
A.M. Adib et al. / Journal of Pharmaceutical and Biomedical Analysis 96 (2014) 104–110
107
Fig. 4. Synchronous 2D-IR contour plot (A), synchronous 2D-IR mesh plot (B) and synchronous 2D-IR auto peaks (C) of ethyl acetate extract (c) and methanol extract (d) (range:1000–1400 cm−1 ).
assigned to the stretching vibration of saturated hydrocarbon groups, the peak at 1619 cm−1 due to the stretching vibration of C=C bonds in compounds embodying aromatic rings, and the peaks at 1236–1036 cm−1 mainly attributed to the stretching vibration of C–O, which displayed the characteristic absorptions of glycosides. In addition, the peak at 1651 and 1700 cm−1 are also assigned to the C=O stretching vibrations. These data indicate that different sort of carbonyl compounds with different quantities existed in B. frutescens leaves. Further, the presence of peaks at ∼1600 cm−1 , ∼1500 cm−1 , ∼1450 cm−1 , and ∼1270 cm−1 are the characteristic absorption peaks of flavonoids which represented the skeletal stretching vibration of the aromatic ring A and B of flavonoids and the functional group =C–O–C of C ring of flavones. The above spectral features indicated that B. frutescens leaves contain flavonoids as reported by Makino and Fujimoto [8]. The tentative assignment of B. frutescens IR spectrum is tabulated in Table 1. 3.2. 1D-FTIR spectral analysis of chloroform and aqueous extract Comparison of the IR spectra of chloroform (Fig. 1b) and aqueous extract (Fig. 1e) revealed that these two extracts are noticeably different in terms of compositions. In the aqueous extract (Fig. 1e), the
strong peak at 3391 cm−1 belonging to the stretching vibrations of O–H groups, while in the chloroform extract (Fig. 1b) the intensity of absorption for O–H groups at 3388 cm−1 was lower as compared to the aqueous extract. The absorptions due to saturated hydrocarbons appear at 2928 cm−1 in the chloroform extract (Fig. 1b) while in the aqueous extract (Fig. 1e) it appears at 2936 cm−1 but with extremely low intensity. This clearly indicates that there are more aliphatic compounds in the chloroform extract as compared to the aqueous extract. The main differences between the chloroform and aqueous extracts are that there is a strong carbonyl peaks at 1689 cm−1 in the spectrum of chloroform extract (Fig. 1b). On the contrary, there is only a shoulder peak of carbonyl at 1655 cm−1 in the IR spectrum of the aqueous extract (Fig. 1e). These data indicate that the amount of carbonyl compounds in the chloroform extract are more than those in the methanol extract owing to the different polarities of extraction solvents. Recently, HPLC analysis on the leaves of B. frutescens has been performed [8]. It has been reported that a flavonoid glycoside compound identified as myricetin 3-O-␣-l-rhamnopyranoside (1) were present in abundance in the leaves of B. frutescens. This flavonoid compound along with chromones such as 8--C-glucopyranosyl5.7-dihydroxy-2-isopropylchromone (2) were detected in the HPLC analysis [19]. In the chloroform (Fig. 2) extract (Fig. 1b), the
108
A.M. Adib et al. / Journal of Pharmaceutical and Biomedical Analysis 96 (2014) 104–110
presence of an absorption band at 1689 cm−1 is the characteristic absorption for conjugated carbonyl bonds. 3.3. Analysis of IR spectra, second derivative spectra and 2D-IR spectra of the methanol extract and ethyl acetate extract Comparison of the ethyl acetate extract (Fig. 1c) and methanol extract (Fig. 1d) reveals that the main constituents in the two extracts seem to be similar. The holistic characters of the strong peaks in the FT-IR spectra indicated that the main components in the two extracts are chromones and flavonoids. In the carbonyl region (1600–1700 cm−1 ) both extracts show three peaks at ∼1700, ∼1657 and ∼1613 cm−1 with almost similar intensity. The peak at ∼1657 cm−1 is often associated to the carbonyl in chromones. This assumption seems to be true when we compared the value of C=O vibration absorption for chromones previously isolated from B. frutescens by Tsui [6] and Satake [7]. Both extracts show similar absorptions at ∼1657, ∼1613, ∼1450, ∼1235, ∼1204, ∼1038, ∼814 and ∼765 cm−1 . However, by comparing the positions, intensities and shapes of the peaks in FT-IR spectra reveals that the chemical constituents in the two extracts are quite different. To begin with, in the IR spectrum of ethyl acetate extract (Fig. 1c), the peak assigned to the stretching vibration of
C–H of CH2 appears at 2930 cm−1 , whereas the similar peak in
that of methanol extract (Fig. 1d) is at 2929 cm−1 with a stronger intensity. The difference in intensity tells us that the amount of methylene in methanol extract (Fig. 1d) is more than that of ethyl acetate extract (Fig. 1c) and the peaks of methylene might be due to the presence of sesquiterpenes. This is in agreement with a related study by Tsui et al., which has reported that the chemical elements of leaves of B. frutescens are sesquiterpenes such as humulene epoxide, eryophyllene epoxide and clovane-2,9-diol [6]. At this point, the tiny differences make it difficult to identify them unambiguously by utilizing the conventional FT-IR technique only. The second derivative spectra can enhance the apparent resolution and amplify even tiny differences in ordinary FT-IR spectra. By using this technique, some overlapped absorption peaks can be resolved. Fig. 3 shows the second derivative infrared spectra of the two extracts in the range of 1000–1800 cm−1 , which contains the main absorption bands of the chemical constituents in the extracts. As the spectra show, the differences between the two extracts can be seen further clearly. First, the intensity of the peak at 1176 cm−1 in the methanol extract is higher as compared to that of ethyl acetate extract at 1177 cm−1 . A distinctive sharp absorption peak at 1558 cm−1 which occurred in the methanol extract was a characteristic feature to distinguish between methanol and
Fig. 5. Synchronous 2D-IR contour plot (A), synchronous 2D-IR mesh plot (B) and synchronous 2D-IR auto peaks (C) of ethyl acetate extract (c) and methanol extract (d) (range:1400–1800 cm−1 ).
A.M. Adib et al. / Journal of Pharmaceutical and Biomedical Analysis 96 (2014) 104–110
ethyl acetate extracts. There is no evidence of characteristic peaks at 1484 cm−1 in the acetate extract. Hence, these two extracts can be identified easily by those enlarged fingerprint characters in the secondary derivative IR spectrum. The 2D-IR correlation analysis can enhance the resolution of a spectrum just through the differences in responses of different submolecular groups in each component of the system to a given external perturbation [14]. Besides it can obtain much new information, which cannot be acquired from conventional IR and its second derivative spectra. Therefore, there is the unique advantage in using 2D-IR correlation spectra to identify and discriminate of complex systems. 2D-IR based on thermal perturbation reveal characteristics behaviours of individual molecular substituents during the temperature increase. This unique technique has been useful in discriminating very complex mixtures such as Chinese medicine [20,21]. By taking advantage of the different of response for different chemical group under external thermal perturbation, we obtained the 2D-IR correlation spectra of the methanol and ethyl acetate extract in order to interpret the differences in between the two
1.15
extracts. In order to obtain enhanced spectral resolution, we carried out synchronous 2D-IR spectroscopy under thermal perturbation from 50 to 120 ◦ C. All the spectra were subjected to baseline automatic correction and the lowest points of the smooth range of 3100–2750 cm−1 and 1760–1650 cm−1 of all spectra were set to be zero (Figs. 4 and 5). The temperature dependant dynamic IR spectra of the two extracts were obtained at temperatures from 50 to 120 ◦ C at 10 ◦ C interval. As illustrated in Fig. 6, all the spectra were overlaid and baseline corrected in the region 3600–2800 cm−1 and 1750–1550 cm−1 . In both extracts, the absorbance of the wide bands at 3325 and 3409 cm−1 (O–H stretching region) decrease with an increase of the temperature. This suggests that higher temperature eliminates the water molecules of the band at 3600–3100 cm−1 . The spectra show that in the range of 3100–2750 cm−1 , the peaks are sensitive to thermal perturbation when heat is applied. The peaks of C–H of the saturated or/and unsaturated hydrocarbon decrease as the temperature increasing in both the extracts. This indicates that the aliphatic compounds in the two extracts are unstable or can easily volatilize as the temperature increase.
3325
50°C
1.20
50°C
1654
1.0
1609
1.0
0.8
Absorbance (A)
Absorbance (A)
109
2976
0.6
0.8
1704
120°C 0.6
0.4 0.4
120°C
0.3 360 0.0
0.2 3400
3200
Wavenumber
30 00
28 00.0
3600-2800 cm 0.90
170 0
165 0
160 0
1550 .0
Wavenumber (cm-1)
-1
(a)
1750.0
(cm-1)
1750-1550 cm-1
(a) 0.80
340 9
50°C
161 2
293 2
0.8
1655
50°C Absorbance (A)
Absorbance (A)
0.6 0.6
0.4
120°C
169 1
0.4
120°C
0.2
0.2 3600 .0
320 0
Wavenumber
(b)
3600-2800 cm-1
1750 .0
2800 .0
(cm-1)
1700
165 0
Wavenumber
(b)
160 0
1550 .0
(cm-1)
1750-1550 cm-1
Fig. 6. Temperature-dependent FTIR spectra of (a) ethyl actetate and (b) methanol extracts in the region of 3600–2800 cm−1 and 1750–1550 cm−1 over the temperature range from 50 to 120 ◦ C.
110
A.M. Adib et al. / Journal of Pharmaceutical and Biomedical Analysis 96 (2014) 104–110
We analyze the synchronous 2D-IR spectra of the two extracts based on the principles above. The auto peaks have close relation with the intensities of corresponding IR peaks, but the difference can be illustrated more clearly and directly in the 2D-IR spectra. Some differences were observed in the 2D-IR spectra of the samples from the two extracts. 2D-FTIR spectroscopic analysis further differentiates the ethyl acetate and methanol extract. Figs. 5 and 6 present the plane figure, stereo fish net and auto peaks for the two extracts in the range of 1400–1000 and 1800–1400 cm−1 . In the synchronous 2D-IR correlation spectra (Fig. 4) the ethyl acetate extract has 6 × 6 peak clusters, while in the methanol extract has 4 × 4 peak clusters. There are two distinct auto peaks at 1159 and 1189 cm−1 in the synchronous spectrum of the methanol extract within 1400–1000 cm−1 (Fig. 4), while in the ethyl acetate extract there is only one distinct autopeak located at 1349 cm−1 , and the intensity is stronger than that in the methanol extracts. The FT-IR and second derivative spectra of the ethyl acetate and methanol extract within the range 1800–1400 cm−1 are almost identical. The noticeable dissimilarities in the synchronous 2D-IR spectra of the two extracts arise in the range of 1400–1800 cm−1 that mainly exhibits the characteristic absorption of carbonyls and olefinic bonds. Within the range, the methanol extract shows significant autopeaks at 1629, 1659 and 1689 cm−1 , while the ethyl acetate extract shows autopeaks at 1629 and 1660 cm−1 but with stronger intensities as compared to those of methanol extract, which indicates that they have stronger self-correlativity in the increasing temperature process. More difference showed up in the synchronous 2D-IR correlation spectra (Fig. 5), in which the ethyl acetate extract has 6 × 6 peak clusters, while in the methanol extract has 4 × 4 peak clusters. As analyzed above, the high resolution 2D-IR spectra can discern the differences in the two extracts easily, and the results are consistent with that from the second derivative and FT-IR spectra. 4. Conclusion By using FT-IR and the corresponding second derivative spectra and 2D-IR correlation spectra to analyze and compare the difference of categories of chemical constituents in B. frutescens leaves and its different extracts, we realize that even though the spectra of these complicated test samples consist of the overlapped peaks of many chemical compositions, the main constituents in them can be identified according to the fingerprint characters, included in intensities, positions, numbers and shapes of the absorptions peaks and the overall contour of FT-IR spectra. It is a known fact that the spectral differences are the reflections of the different chemical constituents in the tested samples. In general, it has been proved that the FT-IR spectrum combined with second derivative spectrum and 2D-IR spectrum is a direct, fast, non destructive and effective method to discriminate samples with similar IR fingerprint features from one another.
Acknowledgement This study was supported by a grant from the Ministry of Agriculture Malaysia grant no. 05-03-10-SF1016. References [1] Herbal Medicine Research Centre, Compendium of Medicinal Plants Used in Malaysia, 1, Herbal Medicine Research Centre, Institute for Medical Research, Kuala Lumpur, 2002, pp. 98–99. [2] Y. Fujimoto, S. Usui, M. Makino, M. Sumatra, Phloroglucinols from Baeckea frutescens, Phytochemistry 41 (1996) 923–925. [3] J.-K. Hwang, J.-S. Shim, J.-Y. Chung, Anticariogenic activity of some tropical medicinal plants against Streptococcus mutans, Fitoterapia 75 (2004) 596–598. [4] T. Murningsih, H. Subekti, K. Matsuraa, M. Takahasi, O. Yamasaki, Y. Yamato, K. Maede, M. Katakura, S. Kobayashi, T. Yushihara, Evaluation of the inhibitory activities of the extracts of Indonesia traditional medicinal plants against Plasmodium falciparum and Babesia gibsoni, J. Vet. Med. Sci. 67 (2005) 829–831. [5] I. Jantan, A.S. Ahmad, S.A. Abu Bakar, A.R. Ahmad, M. Trockenbrodt, C.V. Chak, Constituents of the essential oil of Baeckea frutescens L. from Malaysia, Flavour Fragrance J. 13 (1998) 245–247. [6] W.-Y. Tsui, G.D. Brown, Chromones and chromanones from Baeckea frutescens, Phytochemistry 43 (1996) 871–876. [7] T. Satake, K. Kamiya, Y. Saiki, T. Hama, H. Endang, M. Umar, Chromone Cglycosides from Baeckea frutescens, Phytochemistry 50 (1999) 303–306. [8] M. Makino, Y. Fujimoto, Flavanones from Baeckea frutescens, Phytochemistry 50 (1998) 273–277. [9] Y.W. Wu, S.Q. Sun, J. Zhao, L. Yi, Q. Zhou, Rapid discrimination of extracts of Chinese propolis and poplar buds by FT-IR and 2D IR correlation spectroscopy, J. Mol. Struct. 833–834 (833) (2008) 48–50. [10] C.H. Xu, S.Q. Sun, C.Q. Guo, Q. Zhou, J.X. Tao, I. Noda, Multi-steps infrared macrofingerprint analysis for thermal processing of Fructus viticis, Vib. Spectrosc. 41 (2006) 118–125. [11] L. Zuo, S.Q. Sun, Q. Zhou, J.X. Tao, I. Noda, 2D-IR correlation analysis of deteriorative process of traditional Chinese medicine ‘Qing Kai Ling’ injection, J. Pharm. Biomed. Anal. 30 (2003) 1491–1498. [12] I. Noda, Two-dimensional infrared (2D IR) spectroscopy, Bull. Am. Phys. Soc. 31 (1986) 520–524. [13] I. Noda, Two-dimensional infrared spectroscopy, J. Am. Chem. Soc. 111 (1989) 8116–8119. [14] I. Noda, Generalized two-dimensional correlation method applicable to infrared. Raman and other types of spectroscopy, Appl. Spectrosc. 47 (1993) 1329–1336. [15] A. Nabet, M. Pezolet, Two-dimensional FT-IR spectroscopy: a powerful method to study the secondary structure of proteins using H–D exchange, Appl. Spectrosc. 51 (1997) 466–469. [16] G. Tian, Q. Wu, S. Sun, I. Noda, G. Guo, Study of thermal melting behavior of microbial polyhydroxyalkanoate using two-dimensional Fourier transform infrared correlation spectroscopy, Appl. Spectrosc. 55 (2001) 888–893. [17] Q. Wu, G. Tian, S. Sun, I. Noda, G. Chen, Study of microbial polyhydroxyalkanoates using two-dimensional Fourier-transform infrared correlation spectroscopy, Appl. Polym. Sci. 82 (2001) 934–940. [18] J.X. Xie, J.B. Chang, X.M. Wang, Application of Infrared Spectroscopy in Organic and Pharmaceutical Chemistry, 2nd ed., Science Press, Beijing, 2001. [19] K. Kamiya, T. Satake, Chemical constituents of Baeckea frutescens leaves inhibit copper-induced low-density lipoprotein oxidation, Fitoterapia 81 (2010) 185–189. [20] X. YaO, Y. Peng, Q. Zhou, S.Q. Sun, Distinction of eight Lycium species by Fourier-transform infrared spectroscopy and two-dimensional correlation spectroscopy, J. Mol. Struct. 974 (2010) 161–164. [21] C.-H. Xu, S.Q. Sun, C.-Q. Guo, Q. Zhou, J.-X. Tao, I. Noda, Multi-steps infrared macro fingerprinting analysis for thermal processing Fructus viticis, Vib. Spectrosc. 41 (2006) 118–125.
Contents lists available at ScienceDirect
Journal of Pharmaceutical and Biomedical Analysis journal homepage: www.elsevier.com/locate/jpba
Two-dimensional correlation infrared spectroscopy applied to analyzing and identifying the extracts of Baeckea frutescens medicinal materials Adiana Mohamed Adib ∗ , Fadzureena Jamaludin, Ling Sui Kiong, Nuziah Hashim, Zunoliza Abdullah Natural Products Division, Forest Research Institute Malaysia, 52109 Kepong, Selangor, Malaysia
a r t i c l e
i n f o
Article history: Received 25 September 2013 Received in revised form 13 February 2014 Accepted 15 March 2014 Available online 25 March 2014 Keywords: FT-IR Two-dimensional correlation infrared spectroscopy Second derivative IR spectra Baeckea frutescens
a b s t r a c t Baeckea frutescens or locally known as Cucur atap is used as antibacterial, antidysentery, antipyretic and diuretic agent. In Malaysia and Indonesia, they are used as an ingredient of the traditional medicine given to mothers during confinement. A three-steps infra-red (IR) macro-fingerprinting method combining conventional IR spectra, and the secondary derivative spectra with two dimensional infrared correlation spectroscopy (2D-IR) have been proved to be effective methods to examine a complicated mixture such as herbal medicines. This study investigated the feasibility of employing multi-steps IR spectroscopy in order to study the main constituents of B. frutescens and its different extracts (extracted by chloroform, ethyl acetate, methanol and aqueous in turn). The findings indicated that FT-IR and 2D-IR can provide many holistic variation rules of chemical constituents. The structural information of the samples indicated that B. frutescens and its extracts contain a large amount of flavonoids, since some characteristic absorption peaks of flavonoids, such as ∼1600 cm−1 , ∼1500 cm−1 , ∼1450 cm−1 , and ∼1270 cm−1 can be observed. The macroscopical fingerprint characters of FT-IR and 2D-IR spectra can not only provide the information of main chemical constituents in medicinal materials and their different extracts, but also compare the components differences among the similar samples. In conclusion, the multi-steps IR macro-fingerprint method is rapid, effective, visual and accurate for pharmaceutical research. © 2014 Elsevier B.V. All rights reserved.
1. Introduction Baeckea frutescens or locally known as Cucur atap is a small tree which is found in Peninsular Malaysia and Sumatra. It is also found distributed along the coastal areas of Southern China and Australia. It grows wild on arid soils in the mountain and midlands. In Peninsular Malaysia, it is found both on the mountain and sandy coasts [1]. The tree trunk is short with several upright branches that spread out and droop at the ends as fine twigs. The leaves are opposite, small and very narrow, needle-like, only about 6–15 mm long. In traditional usage, it is claimed to be effective in treating influenza, dyspepsia, jaundice, dysentery, measles and irregular menstrual cycles. Its external usage is in treating furunculosis and impetigo [1]. They are also used in massaging postpartum women for the treatment of body aches and numbness of the
∗ Corresponding author. Tel.: +60 3 62797366/+60 137834263; fax: +60 3 62729805. E-mail addresses: [email protected], [email protected] (A.M. Adib). http://dx.doi.org/10.1016/j.jpba.2014.03.022 0731-7085/© 2014 Elsevier B.V. All rights reserved.
limbs [1]. B. frutescens leaf extract has been reported to have various pharmacological activity such as cytotoxic [2], anticariogenic [3] and antibabesial activities [4]. Chemical studies on the leaves of B. frutescens have indicated the presence of volatile oil [5], sesquiterpenes [6], chromone C-glycosides [7], phloroglucinols [2] and flavanones [8]. Currently, the main method for identifying the qualities of medicinal materials and extract products is to use chromatography examining the content of certain chemical constituent in the tested sample. It is well known that medicinal materials comprise hundreds of components, and produce their curative effects through synergistic effects of many ingredients, so the limited numbers of specific components cannot availably reflect the real qualities of the herbal medicines. Therefore, we need to find a quick and effective analysis to entirely monitor and reflect the whole constituents of the medicinal materials and their corresponding extract products. In recent years, there are many reports on the application of FT-IR in medicinal plants research on various aspects such as identification, quality control and forecasting stability [9–11]. For a complex system, FT-IR has shown advantages over the other
A.M. Adib et al. / Journal of Pharmaceutical and Biomedical Analysis 96 (2014) 104–110
105
1619
3374
1036 1449 1376 1236 1517 1161 1542 1319 1106
2926 2870
616 764
(a)
2928
819 1689
3388
2871
1238 1616
3352
1377 1138 1456 1274 1167 1386 1036 1189 1104 1316 1515
1613 1657
Absorbance (A)
997 975 889
(b)
2978
724 662 608 828 766 570
464
1202 1449 1350
1235
1501 1536
1704
1167
1038 1086 916 966
3398
667
814
717 643 764 678 576
(c) 1614 2929
1657
1450
1204 1380
1235 1142
1690
1039
1515 817
3391
765
612
(d)
1614 2936
1449
1374
1236
1070 1045
1515
915
(e)
4000.0
3600
3200
2800
2400
2000
1800
1600
1400
1200
1000
768 818
800
716
614
600
450.0
Wavenumber (cm-1)
Fig. 1. FT-IR spectra of B. frutescens and its different extracts: (a) B. frutescens; (b) dichloromethane extract; (c) ethyl acetate extract; (d) methanol extract; (e) aqueous extract.
conventional analysis methods, as it presents the whole features as well as the macro-fingerprint characters of the sample. Owing to the fingerprint characters and extensive applicability to the samples, FT-IR has played an important role in pharmaceutical analysis in recent years. Two-dimensional correlation infrared (2DIR) spectroscopy was proposed by Noda [12]. The construction of 2D correlation IR spectra is based on the detection of dynamic changes of a system under an external perturbation. Such a perturbation induces selective changes in molecular constituents associated with individual normal modes of vibration in the system. The correlation analysis is applied to a set of spectra taken during the perturbation, so as to yield 2D correlation IR spectra. There has been a rapid progress in the technique of 2D correlation spectral analysis. The external perturbations used to produce 2D correlation IR spectra were sinusoids in early days [13]. Nowadays, however, arbitrary wave forms with non-periodical perturbations are commonly used [14]. The applied perturbations can be light, heat, electricity, magnetism, chemistry, concentration changes, or mechanical force [15–17]. As long as the samples are different in components or contents, the differences will embody in their FT-IR spectra. In this paper, we report the main components and the holistic variation rules of chemical constituents in B. frutescens and its extracts by applying FT-IR, second derivative infrared and
2D-IR spectroscopy. The purpose of this study is to develop a rapid, accurate and feasible analytical method to appraise integrally the inherent qualities of herbal material and its corresponding extracts. 2. Experimental 2.1. Apparatus IR spectra were recorded on a Spectrum 100 Fourier transform infrared (FTIR) spectrometer (Perkin-Elmer, Santa Clara, CA, USA), equipped with a mid-infrared deuterated triglycine sulfate (DTGS) detector. The spectra were obtained from the scan range of 4000 to 450 cm−1 with a resolution of 4 cm−1 and with a total accumulation of 16 scans. Portable programmable temperature controller (4000 series TM High Stability Temperature Controller, Specac, Ltd.) was used in the range of 50–120 ◦ C. 2.2. Plant material Samples of B. frutescens were collected in January 2010 from Forest Research Institute Malaysia (FRIM) Research Station in Setiu, Terengganu, Malaysia. The plant was identified at Biological Resources Programme FRIM and the voucher specimen was
Table 1 The tentative assignment of Baeckea frutescen IR spectrum [18]. Band (cm−1 )
Base group and vibration mode
Main attribution
3374 2926 2870 1619 1517 1449 1319 1236 1106 1036
(O–H) as (C–H) s (C–H) (C=O), (C=C) rf (ar), rf (ha) rf (ar), rf (ha) I,as (=C–O–C) ıs (C–CO), (C–O), II,as (=C–O–C), f (C–C) ı(C–OH), f (C–C) ı(C–OH), I (=C–O–C), f (C–C)
OH CH3 , CH2 CH3 , CH2 Flavonoids, chromones, phlororoglucinols Flavonoids, chromones, phlororoglucinols Flavonoids, chromones, phlororoglucinols Flavonoids, chromones, phlororoglucinols Flavanoids, chromones phlororoglucinols, branched chain hydrocarbon Compound which has secondary alcohol functional group Flavonoids, compound which has primary alcohol functional group
Note: , stretching or frame vibration; ı, bending; s, symmetrical; as, asymmetrical; ar, aromatic, ha, heteroaromatic; rf, ring frame; f, frame; I, II, stretching mode.
106
A.M. Adib et al. / Journal of Pharmaceutical and Biomedical Analysis 96 (2014) 104–110
measurement of 2D-FTIR spectra, each sample disc was put into the sample pool connected with a temperature controller. Dynamic spectra were collected at different temperatures ranging from 50 to 120 ◦ C at an interval of 10 ◦ C. 2D-IR correlation spectra were acquired by treatment of the series of temperature-dependant dynamic spectra with 2D-IR correlation analysis employing Softdoc software developed by Tsinghua University (Beijing, China).
OH OH HO
Glc HO
O
O
OH O Rha
OH O
OH O (1)
(2)
3. Results and discussion
Fig. 2. Myricetin 3-O-␣-L-rhamnopyranoside (1) and 8--C-glucopyranosyl-5,7dihydroxy-2-isopropylchromone (2).
3.1. 1D-FTIR spectral analysis of B. frutescens leaves Fig. 1 shows the FT-IR spectra of B. frutescens leaves and its different extract at room temperature. The leaves of B. frutescen and its different extracts are complex mixtures which showed IR spectra with overlapping absorption peaks. In the spectra, we can see clearly that although they show substantial overlap of each absorption spectrum of various components, each band represents an overall overlap of some characteristic absorption peaks of functional groups in the samples. The FT-IR spectra of ethyl acetate extract (c) and methanol extract (d) are rather similar, but obviously differ from the spectra of B. frutescens leaves (a), chloroform extract (b) and aqueous extract (e). Considering B. frutescens leaves (Fig. 1a) is a mixture of complex compounds, its infrared spectrum is complex and contains several bands from the contribution of different functional groups. As expected the FT-IR spectrum shows a total overlap of each absorption spectrum of various components. The chemical compounds of B. frutescens leaves are sesquiterpenes [6], chromone C-glycosides [7], phloroglucinols [2] and flavonoids [8]. The principal components in B. frutescens leaves all have their own infrared characteristic peaks. As a kind of macro-fingerprinting features of natural product complex, several characters can be extracted such as the strongest peak at 3374 cm−1 belonging to the stretching vibration of O–H groups, the peak 2926 cm−1 and 2870 cm−1
deposited at the Natural Products Division Reference Collection, FRIM Kepong. The leaves of the plants were removed and used for macro-fingerprinting analysis. 2.3. Procedure The dried leaves of B. frutescens (2 kg) were pulverized into coarse powder. One-tenth of the coarse powder was further ground into fine powder form (sample a). The remaining coarse powder was orderly extracted by chloroform, ethyl acetate, methanol and water for three times, respectively. All the extracts were concentrated under vacuum and dried to yield the following samples; (b) chloroform (19.2 g); (c) ethyl acetate (27.4 g) and (d) methanol (35.7 g) and aqueous (31.2 g) extracts. After that, exactly 2 mg of each sample was mixed with 100 mg of potassium bromide (KBr) and the mixture was further ground and pressed under 10 t of pressure to produce a thin disc with 13 mm of diameter. 1D-FTIR spectra were recorded from a total of 16 scans in the 4000 to 450 cm−1 range with a resolution of 4 cm−1 . The second derivative spectra were obtained by using Savitzky–Golay filter through 13 point smoothing. Savitzky–Golay smoothing aimed for minimum distortion by least squares fitting a cubic polynomial. For the 1673 1639
1176 1558
1459
1720
1542
1628 1619
1698
1013
1484
1586
1735 1753
1787
1050 1152
1523
1434 1418
12781260
1409
1367
1215
1065
1333
1506
11241105
1299
1078
1029
1001
1710
Absorbance (A)
1449
1677
1642 1153
1478 1753
1786
1765
1734
1523 1576 15631553
1704
1538
1420 1435
1505
1776
1225
1394 1372 1336 1360
1177
12651246 1279
1051 1066
1013
1110 1137
1459
1030
1616
1800.0
1750
1700
1650
1600
1550
1500
1450
1400
1350
1300
1250
1200
1150
Wavenumber (cm-1)
Fig. 3. Second derivative spectra of the ethyl acetate extract and methanol extract.
1100
1050
1000.0
A.M. Adib et al. / Journal of Pharmaceutical and Biomedical Analysis 96 (2014) 104–110
107
Fig. 4. Synchronous 2D-IR contour plot (A), synchronous 2D-IR mesh plot (B) and synchronous 2D-IR auto peaks (C) of ethyl acetate extract (c) and methanol extract (d) (range:1000–1400 cm−1 ).
assigned to the stretching vibration of saturated hydrocarbon groups, the peak at 1619 cm−1 due to the stretching vibration of C=C bonds in compounds embodying aromatic rings, and the peaks at 1236–1036 cm−1 mainly attributed to the stretching vibration of C–O, which displayed the characteristic absorptions of glycosides. In addition, the peak at 1651 and 1700 cm−1 are also assigned to the C=O stretching vibrations. These data indicate that different sort of carbonyl compounds with different quantities existed in B. frutescens leaves. Further, the presence of peaks at ∼1600 cm−1 , ∼1500 cm−1 , ∼1450 cm−1 , and ∼1270 cm−1 are the characteristic absorption peaks of flavonoids which represented the skeletal stretching vibration of the aromatic ring A and B of flavonoids and the functional group =C–O–C of C ring of flavones. The above spectral features indicated that B. frutescens leaves contain flavonoids as reported by Makino and Fujimoto [8]. The tentative assignment of B. frutescens IR spectrum is tabulated in Table 1. 3.2. 1D-FTIR spectral analysis of chloroform and aqueous extract Comparison of the IR spectra of chloroform (Fig. 1b) and aqueous extract (Fig. 1e) revealed that these two extracts are noticeably different in terms of compositions. In the aqueous extract (Fig. 1e), the
strong peak at 3391 cm−1 belonging to the stretching vibrations of O–H groups, while in the chloroform extract (Fig. 1b) the intensity of absorption for O–H groups at 3388 cm−1 was lower as compared to the aqueous extract. The absorptions due to saturated hydrocarbons appear at 2928 cm−1 in the chloroform extract (Fig. 1b) while in the aqueous extract (Fig. 1e) it appears at 2936 cm−1 but with extremely low intensity. This clearly indicates that there are more aliphatic compounds in the chloroform extract as compared to the aqueous extract. The main differences between the chloroform and aqueous extracts are that there is a strong carbonyl peaks at 1689 cm−1 in the spectrum of chloroform extract (Fig. 1b). On the contrary, there is only a shoulder peak of carbonyl at 1655 cm−1 in the IR spectrum of the aqueous extract (Fig. 1e). These data indicate that the amount of carbonyl compounds in the chloroform extract are more than those in the methanol extract owing to the different polarities of extraction solvents. Recently, HPLC analysis on the leaves of B. frutescens has been performed [8]. It has been reported that a flavonoid glycoside compound identified as myricetin 3-O-␣-l-rhamnopyranoside (1) were present in abundance in the leaves of B. frutescens. This flavonoid compound along with chromones such as 8--C-glucopyranosyl5.7-dihydroxy-2-isopropylchromone (2) were detected in the HPLC analysis [19]. In the chloroform (Fig. 2) extract (Fig. 1b), the
108
A.M. Adib et al. / Journal of Pharmaceutical and Biomedical Analysis 96 (2014) 104–110
presence of an absorption band at 1689 cm−1 is the characteristic absorption for conjugated carbonyl bonds. 3.3. Analysis of IR spectra, second derivative spectra and 2D-IR spectra of the methanol extract and ethyl acetate extract Comparison of the ethyl acetate extract (Fig. 1c) and methanol extract (Fig. 1d) reveals that the main constituents in the two extracts seem to be similar. The holistic characters of the strong peaks in the FT-IR spectra indicated that the main components in the two extracts are chromones and flavonoids. In the carbonyl region (1600–1700 cm−1 ) both extracts show three peaks at ∼1700, ∼1657 and ∼1613 cm−1 with almost similar intensity. The peak at ∼1657 cm−1 is often associated to the carbonyl in chromones. This assumption seems to be true when we compared the value of C=O vibration absorption for chromones previously isolated from B. frutescens by Tsui [6] and Satake [7]. Both extracts show similar absorptions at ∼1657, ∼1613, ∼1450, ∼1235, ∼1204, ∼1038, ∼814 and ∼765 cm−1 . However, by comparing the positions, intensities and shapes of the peaks in FT-IR spectra reveals that the chemical constituents in the two extracts are quite different. To begin with, in the IR spectrum of ethyl acetate extract (Fig. 1c), the peak assigned to the stretching vibration of
C–H of CH2 appears at 2930 cm−1 , whereas the similar peak in
that of methanol extract (Fig. 1d) is at 2929 cm−1 with a stronger intensity. The difference in intensity tells us that the amount of methylene in methanol extract (Fig. 1d) is more than that of ethyl acetate extract (Fig. 1c) and the peaks of methylene might be due to the presence of sesquiterpenes. This is in agreement with a related study by Tsui et al., which has reported that the chemical elements of leaves of B. frutescens are sesquiterpenes such as humulene epoxide, eryophyllene epoxide and clovane-2,9-diol [6]. At this point, the tiny differences make it difficult to identify them unambiguously by utilizing the conventional FT-IR technique only. The second derivative spectra can enhance the apparent resolution and amplify even tiny differences in ordinary FT-IR spectra. By using this technique, some overlapped absorption peaks can be resolved. Fig. 3 shows the second derivative infrared spectra of the two extracts in the range of 1000–1800 cm−1 , which contains the main absorption bands of the chemical constituents in the extracts. As the spectra show, the differences between the two extracts can be seen further clearly. First, the intensity of the peak at 1176 cm−1 in the methanol extract is higher as compared to that of ethyl acetate extract at 1177 cm−1 . A distinctive sharp absorption peak at 1558 cm−1 which occurred in the methanol extract was a characteristic feature to distinguish between methanol and
Fig. 5. Synchronous 2D-IR contour plot (A), synchronous 2D-IR mesh plot (B) and synchronous 2D-IR auto peaks (C) of ethyl acetate extract (c) and methanol extract (d) (range:1400–1800 cm−1 ).
A.M. Adib et al. / Journal of Pharmaceutical and Biomedical Analysis 96 (2014) 104–110
ethyl acetate extracts. There is no evidence of characteristic peaks at 1484 cm−1 in the acetate extract. Hence, these two extracts can be identified easily by those enlarged fingerprint characters in the secondary derivative IR spectrum. The 2D-IR correlation analysis can enhance the resolution of a spectrum just through the differences in responses of different submolecular groups in each component of the system to a given external perturbation [14]. Besides it can obtain much new information, which cannot be acquired from conventional IR and its second derivative spectra. Therefore, there is the unique advantage in using 2D-IR correlation spectra to identify and discriminate of complex systems. 2D-IR based on thermal perturbation reveal characteristics behaviours of individual molecular substituents during the temperature increase. This unique technique has been useful in discriminating very complex mixtures such as Chinese medicine [20,21]. By taking advantage of the different of response for different chemical group under external thermal perturbation, we obtained the 2D-IR correlation spectra of the methanol and ethyl acetate extract in order to interpret the differences in between the two
1.15
extracts. In order to obtain enhanced spectral resolution, we carried out synchronous 2D-IR spectroscopy under thermal perturbation from 50 to 120 ◦ C. All the spectra were subjected to baseline automatic correction and the lowest points of the smooth range of 3100–2750 cm−1 and 1760–1650 cm−1 of all spectra were set to be zero (Figs. 4 and 5). The temperature dependant dynamic IR spectra of the two extracts were obtained at temperatures from 50 to 120 ◦ C at 10 ◦ C interval. As illustrated in Fig. 6, all the spectra were overlaid and baseline corrected in the region 3600–2800 cm−1 and 1750–1550 cm−1 . In both extracts, the absorbance of the wide bands at 3325 and 3409 cm−1 (O–H stretching region) decrease with an increase of the temperature. This suggests that higher temperature eliminates the water molecules of the band at 3600–3100 cm−1 . The spectra show that in the range of 3100–2750 cm−1 , the peaks are sensitive to thermal perturbation when heat is applied. The peaks of C–H of the saturated or/and unsaturated hydrocarbon decrease as the temperature increasing in both the extracts. This indicates that the aliphatic compounds in the two extracts are unstable or can easily volatilize as the temperature increase.
3325
50°C
1.20
50°C
1654
1.0
1609
1.0
0.8
Absorbance (A)
Absorbance (A)
109
2976
0.6
0.8
1704
120°C 0.6
0.4 0.4
120°C
0.3 360 0.0
0.2 3400
3200
Wavenumber
30 00
28 00.0
3600-2800 cm 0.90
170 0
165 0
160 0
1550 .0
Wavenumber (cm-1)
-1
(a)
1750.0
(cm-1)
1750-1550 cm-1
(a) 0.80
340 9
50°C
161 2
293 2
0.8
1655
50°C Absorbance (A)
Absorbance (A)
0.6 0.6
0.4
120°C
169 1
0.4
120°C
0.2
0.2 3600 .0
320 0
Wavenumber
(b)
3600-2800 cm-1
1750 .0
2800 .0
(cm-1)
1700
165 0
Wavenumber
(b)
160 0
1550 .0
(cm-1)
1750-1550 cm-1
Fig. 6. Temperature-dependent FTIR spectra of (a) ethyl actetate and (b) methanol extracts in the region of 3600–2800 cm−1 and 1750–1550 cm−1 over the temperature range from 50 to 120 ◦ C.
110
A.M. Adib et al. / Journal of Pharmaceutical and Biomedical Analysis 96 (2014) 104–110
We analyze the synchronous 2D-IR spectra of the two extracts based on the principles above. The auto peaks have close relation with the intensities of corresponding IR peaks, but the difference can be illustrated more clearly and directly in the 2D-IR spectra. Some differences were observed in the 2D-IR spectra of the samples from the two extracts. 2D-FTIR spectroscopic analysis further differentiates the ethyl acetate and methanol extract. Figs. 5 and 6 present the plane figure, stereo fish net and auto peaks for the two extracts in the range of 1400–1000 and 1800–1400 cm−1 . In the synchronous 2D-IR correlation spectra (Fig. 4) the ethyl acetate extract has 6 × 6 peak clusters, while in the methanol extract has 4 × 4 peak clusters. There are two distinct auto peaks at 1159 and 1189 cm−1 in the synchronous spectrum of the methanol extract within 1400–1000 cm−1 (Fig. 4), while in the ethyl acetate extract there is only one distinct autopeak located at 1349 cm−1 , and the intensity is stronger than that in the methanol extracts. The FT-IR and second derivative spectra of the ethyl acetate and methanol extract within the range 1800–1400 cm−1 are almost identical. The noticeable dissimilarities in the synchronous 2D-IR spectra of the two extracts arise in the range of 1400–1800 cm−1 that mainly exhibits the characteristic absorption of carbonyls and olefinic bonds. Within the range, the methanol extract shows significant autopeaks at 1629, 1659 and 1689 cm−1 , while the ethyl acetate extract shows autopeaks at 1629 and 1660 cm−1 but with stronger intensities as compared to those of methanol extract, which indicates that they have stronger self-correlativity in the increasing temperature process. More difference showed up in the synchronous 2D-IR correlation spectra (Fig. 5), in which the ethyl acetate extract has 6 × 6 peak clusters, while in the methanol extract has 4 × 4 peak clusters. As analyzed above, the high resolution 2D-IR spectra can discern the differences in the two extracts easily, and the results are consistent with that from the second derivative and FT-IR spectra. 4. Conclusion By using FT-IR and the corresponding second derivative spectra and 2D-IR correlation spectra to analyze and compare the difference of categories of chemical constituents in B. frutescens leaves and its different extracts, we realize that even though the spectra of these complicated test samples consist of the overlapped peaks of many chemical compositions, the main constituents in them can be identified according to the fingerprint characters, included in intensities, positions, numbers and shapes of the absorptions peaks and the overall contour of FT-IR spectra. It is a known fact that the spectral differences are the reflections of the different chemical constituents in the tested samples. In general, it has been proved that the FT-IR spectrum combined with second derivative spectrum and 2D-IR spectrum is a direct, fast, non destructive and effective method to discriminate samples with similar IR fingerprint features from one another.
Acknowledgement This study was supported by a grant from the Ministry of Agriculture Malaysia grant no. 05-03-10-SF1016. References [1] Herbal Medicine Research Centre, Compendium of Medicinal Plants Used in Malaysia, 1, Herbal Medicine Research Centre, Institute for Medical Research, Kuala Lumpur, 2002, pp. 98–99. [2] Y. Fujimoto, S. Usui, M. Makino, M. Sumatra, Phloroglucinols from Baeckea frutescens, Phytochemistry 41 (1996) 923–925. [3] J.-K. Hwang, J.-S. Shim, J.-Y. Chung, Anticariogenic activity of some tropical medicinal plants against Streptococcus mutans, Fitoterapia 75 (2004) 596–598. [4] T. Murningsih, H. Subekti, K. Matsuraa, M. Takahasi, O. Yamasaki, Y. Yamato, K. Maede, M. Katakura, S. Kobayashi, T. Yushihara, Evaluation of the inhibitory activities of the extracts of Indonesia traditional medicinal plants against Plasmodium falciparum and Babesia gibsoni, J. Vet. Med. Sci. 67 (2005) 829–831. [5] I. Jantan, A.S. Ahmad, S.A. Abu Bakar, A.R. Ahmad, M. Trockenbrodt, C.V. Chak, Constituents of the essential oil of Baeckea frutescens L. from Malaysia, Flavour Fragrance J. 13 (1998) 245–247. [6] W.-Y. Tsui, G.D. Brown, Chromones and chromanones from Baeckea frutescens, Phytochemistry 43 (1996) 871–876. [7] T. Satake, K. Kamiya, Y. Saiki, T. Hama, H. Endang, M. Umar, Chromone Cglycosides from Baeckea frutescens, Phytochemistry 50 (1999) 303–306. [8] M. Makino, Y. Fujimoto, Flavanones from Baeckea frutescens, Phytochemistry 50 (1998) 273–277. [9] Y.W. Wu, S.Q. Sun, J. Zhao, L. Yi, Q. Zhou, Rapid discrimination of extracts of Chinese propolis and poplar buds by FT-IR and 2D IR correlation spectroscopy, J. Mol. Struct. 833–834 (833) (2008) 48–50. [10] C.H. Xu, S.Q. Sun, C.Q. Guo, Q. Zhou, J.X. Tao, I. Noda, Multi-steps infrared macrofingerprint analysis for thermal processing of Fructus viticis, Vib. Spectrosc. 41 (2006) 118–125. [11] L. Zuo, S.Q. Sun, Q. Zhou, J.X. Tao, I. Noda, 2D-IR correlation analysis of deteriorative process of traditional Chinese medicine ‘Qing Kai Ling’ injection, J. Pharm. Biomed. Anal. 30 (2003) 1491–1498. [12] I. Noda, Two-dimensional infrared (2D IR) spectroscopy, Bull. Am. Phys. Soc. 31 (1986) 520–524. [13] I. Noda, Two-dimensional infrared spectroscopy, J. Am. Chem. Soc. 111 (1989) 8116–8119. [14] I. Noda, Generalized two-dimensional correlation method applicable to infrared. Raman and other types of spectroscopy, Appl. Spectrosc. 47 (1993) 1329–1336. [15] A. Nabet, M. Pezolet, Two-dimensional FT-IR spectroscopy: a powerful method to study the secondary structure of proteins using H–D exchange, Appl. Spectrosc. 51 (1997) 466–469. [16] G. Tian, Q. Wu, S. Sun, I. Noda, G. Guo, Study of thermal melting behavior of microbial polyhydroxyalkanoate using two-dimensional Fourier transform infrared correlation spectroscopy, Appl. Spectrosc. 55 (2001) 888–893. [17] Q. Wu, G. Tian, S. Sun, I. Noda, G. Chen, Study of microbial polyhydroxyalkanoates using two-dimensional Fourier-transform infrared correlation spectroscopy, Appl. Polym. Sci. 82 (2001) 934–940. [18] J.X. Xie, J.B. Chang, X.M. Wang, Application of Infrared Spectroscopy in Organic and Pharmaceutical Chemistry, 2nd ed., Science Press, Beijing, 2001. [19] K. Kamiya, T. Satake, Chemical constituents of Baeckea frutescens leaves inhibit copper-induced low-density lipoprotein oxidation, Fitoterapia 81 (2010) 185–189. [20] X. YaO, Y. Peng, Q. Zhou, S.Q. Sun, Distinction of eight Lycium species by Fourier-transform infrared spectroscopy and two-dimensional correlation spectroscopy, J. Mol. Struct. 974 (2010) 161–164. [21] C.-H. Xu, S.Q. Sun, C.-Q. Guo, Q. Zhou, J.-X. Tao, I. Noda, Multi-steps infrared macro fingerprinting analysis for thermal processing Fructus viticis, Vib. Spectrosc. 41 (2006) 118–125.
