- Email: [email protected]
Study of traditional Chinese animal drugs using FT-IR and 2D-IR correlation spectroscopy
Spectrochimica Acta Part A 63 (2006) 565–573
Study of traditional Chinese animal drugs using FT-IR and 2D-IR correlation spectroscopy Ying-Ming Li a , Su-Qin Sun a,∗ , Qun Zhou a , Jia-Xun Tao a , Isao Noda b a
Key Laboratory of Bioorganic Phosphorus Chemistry & Chemical Biology (Ministry of Education), Department of Chemistry, Tsinghua University, Beijing 100084, PR China b The Procter & Gamble Co., West Chester, OH 45069, USA Received 28 March 2005; received in revised form 2 June 2005; accepted 3 June 2005
Abstract Using FT-IR and 2D-IR correlation spectroscopy, the intensity changes and their sequence of amide I and II bands of four traditional Chinese animal drugs (Cornu Cervi Pantotrichum, Cornu Saigae Tataricae, Scorpio and Hirud) under thermal perturbation are studied, and component of Ca3 (PO4 )2 in the drug of Cornu Cervi Pantotrichum and sulfates components in the drug of scorpion are identified. The drug of Cornu Cervi Pantotrichum contains inorganic salt Ca3 (PO4 )2 and the drug of Scorpio contains sulfates. It is assigned that the bands of 604 and 561 cm−1 belong to the component of Ca3 (PO4 )2 , and the bands of 637 and 615 cm−1 belong to sulfates. Organic components of these drugs respond to the thermal perturbation far stronger than that of the inorganic components. The intensities of amide I and II bands in the drugs, except for amide II band in drug Scorpio, change strongly. For the drugs of Cornu Cervi Pantotrichum, Cornu 4 Saigae Tataricae and Hirudo, the intensity changes of amide II band occurs prior to that of amide I band. The C N bond in the conjugative system converts from double bond to single bond, but the C O bond remains double bond during the heating process. 3 On the other hand, amide II vibrations, which may involve much more of the hydrogen bonded local structures of amide groups in the system compared to the C O dominated amide I vibration, may undergo thermally induced changes at a temperature much lower than the other mode. The traditional Chinese animal drugs can be identified rapidly and non-separately by using FT-IR and 2D-IR correlation spectroscopy. © 2005 Elsevier B.V. All rights reserved. Keywords: FT-IR; 2D-IR correlation spectroscopy; Animal drugs
1. Introduction Perturbation-based 2D correlation spectroscopy was originally developed in IR field. Compared with conventional spectra, 2D correlation spectra can enhance the spectral resolution by spreading peaks over the second dimension, and simplify the complex spectra consisting of many overlapped peaks, as well as identify various inter and intra molecular interactions through selective correlation of IR peaks [1–3]. The 2D correlation technique has been widely used in traditional spectroscopy, such as IR, Raman, UV and fluorescence. As for 2D-IR correlation spectroscopy, there ∗
Corresponding author. Tel.: +86 10 62787661; fax: +86 10 62770327. E-mail address: [email protected] (S.-Q. Sun).
1386-1425/$ – see front matter © 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.saa.2005.06.004
have been studies on many aspects, including biologically important systems like N-methylacetamide (NMA) [4]. As a non-separate, non-extraction, rapid and convenient method, 2D-IR spectroscopy combined with IR spectroscopy which we called “IR macro-fingerprint method” has extended the application of 2D correlation spectroscopy from simple compounds to very complex systems, such as the field of herbal medicines, including raw herbs [5a], processed medicines, formula granules [5b] and injections [5c]. The IR macro-fingerprint method follows the integrate principle of traditional Chinese medicine, and it does not lose original natural instinct and compatibility of traditional Chinese medicine. This method has been successfully used to the trace and analysis of the deteriorative process of injections, such as Qingkailing injection [6]. It proves that the dete-
Y.-M. Li et al. / Spectrochimica Acta Part A 63 (2006) 565–573
566
rioration of the Qingkailing injection in air is due to the oxidation of flavones and/or the variations in conformation and structure of the flavones compounds. This method also has been applied to the quality control of traditional Chinese medicines. We have made over 200 2D-IR spectra of traditional Chinese medicines and further proposed multistep IR identification method to identify herbal medicines: the IR spectra as the first step identification; the second derivative spectra as the second step identification; the 2D-IR correlation spectra as the third step identification [8]. It presents a new idea for the analysis of complex systems. The so-called animal drugs have reliable and special curative effects and are the important part of traditional Chinese medicine. Their main components contain proteins. The IR spectroscopy is rarely applied directly to the study of animal drugs. In this paper, based on the above method, we emphasized and analyzed the macro-fingerprint features of animal drugs Cornu Cervi Pantotrichum, Cornu Saigae Tataricae, Scorpio and Hirud, as well as their response to thermal perturbations, which is very important to the identification of animal drugs and further to realize their pharmacological efficacies.
The reagents of Ca3 (PO4 )2 and Na2 SO4 are used as received. All animal drugs were identified and provided by the National Institute for the Control of Pharmaceutical and Biological Products of China. 2.2. Procedure Each sample of animal drugs was ground into powder with over 200 meshes and then blended with KBr powder, ground again and pressed into a tablet. After that their IR spectra were collected at room temperature. The dynamic spectra were collected at different temperatures from 50 to 120 ◦ C at interval of 10 ◦ C and were processed by baseline correction. We made baseline autocorrection of spectra during the temperature increasing process before 2D correlation analysis to overcome the baseline drifts. 2D-IR correlation spectra were obtained by the treatment of a series of dynamic spectra with 2D-IR correlation analysis software programmed by our group.
3. Results and discussion 2. Experiment
3.1. IR spectra
2.1. Apparatus and samples
Fig. 1 shows the IR spectra of the four animal drugs. The assignments of characteristic peaks of the spectra are shown in Table 1. There are some similarities of the spectra, such as the strong absorption peaks of the amide bands, since the animal drugs contain a high level of protein components. The amide I bands are assigned to the stretching frequency of C O groups, and the amide II bands are mainly associated
Spectrum GX FT-IR spectrometer (Perkin-Elmer) is equipped with a DTGS detector and portable programmable temperature controller (Model 50–886, Love Control Corporation). IR spectra were recorded from an accumulation of 32 scans in 4000–400 cm−1 range with a resolution of 4 cm−1 .
Fig. 1. The IR spectra of four animal drugs: (a) Cornu Cervi Pantotrichum; (b) Cornu Saigae Tataricae; (c) Scorpion; (d) Hirudo.
Y.-M. Li et al. / Spectrochimica Acta Part A 63 (2006) 565–573
567
Table 1 The assignments of main IR absorption bands of the drugs Animal drug
Amide I (cm−1 )
Amide II (cm−1 )
δ(CH2 , CH3 ) (cm−1 )
γ(CH2 , CH3 ) (cm−1 )
δO–P–O (cm−1 )
δO–S–O (cm−1 )
Cornu Cervi Pantotrichum Cornu Saigae Tataricae Scorpio Hirudo
1659 1657 1662 1659
1546 1517 1531 1537
2956, 2925 2960, 2933 2959, 2927 2959, 2927
2853, 1453 2875, 1451 2855, 1442 2855, 1453
561 – – –
– – 615 –
Fig. 2. Comparison of IR spectra (left) and second derivative spectra (right) of: (a) Cornu Cervi Pantotrichum and (b) Ca3 (PO4 )2 .
with stretching frequency of C N bond, as well as bending vibration of N H groups. However, the positions of these absorption bands related with the amide groups are different for different animal drugs. For example, the characteristic amide bands appear at 1659 and 1546 cm−1 for Cornu Cervi Pantotrichum; at 1657 and 1517 cm−1 for Cornu Saigao; at 1662 and 1531 cm−1 for Scorpion; at 1659 and 1537 cm−1 for Hirudo. Some unique peaks can be observed such as peaks at 1033, 604 and 561 cm−1 for Cornu Cervi Pantotrichum, shown in Fig. 1a and peaks at 1112, 637 and 615 cm−1 for Scorpion, shown in Fig. 1c. We measured the IR spectra and the second derivative spectra of Ca3 (PO4 )2 and Na2 SO4 , and compared these spectra with the spectra of nearly fifty kinds of inorganic compounds including Mg3 (PO4 )2 , Na3 PO4 , etc., and found that in the range of 1100–400 cm−1 the spectra of Cornu Cervi Pantotrichum is very similar to that of Ca3 (PO4 )2 : the peaks at 1033, 961, 604 and 561 cm−1 in IR spectrum of Cornu Cervi Pantotrichum correspond to
the absorption bands at 1031, 962, 603 and 564 cm−1 of Ca3 (PO4 )2 and the peaks at 1033, 961, 604 and 561 cm−1 in the second derivative spectrum correspond to the absorption bands at 1030, 961, 602 and 563 cm−1 of Ca3 (PO4 )2 , shown in Fig. 2, and the spectra of Scorpio is very similar to that of sulfate in the range of 1200–400 cm−1 : the peaks at 1112, 637 and 615 cm−1 in the IR spectrum of Scorpio correspond to the peaks at 1118, 638 and 618 cm−1 of Na2 SO4 and the peaks at 638 and 614 cm−1 in the second derivative spectra of Scorpio correspond to the absorption bands at 639 and 618 cm−1 of Na2 SO4 , shown in Fig. 3. In general, the absorption bands of PO4 3− appear in ranges within 1120–940 cm−1 and within 650–540 cm−1 , and the peak at 567 cm−1 is assigned to bending absorption band of O P O for PO4 3− [9,10]. Zhen et al. [11] considered that in the spectrum of the drug Cornu Cervi Pantotrichum, the peaks below 1100 cm−1 is related to the C C stretching frequencies with medium intensities and the O P O vibrations
Fig. 3. Comparison of IR spectra (left) and second derivative spectra (right) of: (a) Scorpio and (b) Na2 SO4 .
568
Y.-M. Li et al. / Spectrochimica Acta Part A 63 (2006) 565–573
of the calcium phosphate components. The peaks in the range of 1100–400 cm−1 in Fig. 2a are in agreement with the characteristic absorption bands of bone black, which contains 90–94% calcium phosphate and 6–10% carbon [12]. So, it is concluded that the animal drug Cornu Cervi Pantotrichum contain much Ca3 (PO4 )2 components, of which the peaks at 604 and 561 cm−1 (the O P O bending vibration) are characteristic absorption bands. The bending vibrations of O S O in SO4 2− often appear at range within 960–1030 cm−1 with strong and sharp peak and at range within 570–680 cm−1 with medium intensity. And the peak at 615 cm−1 has been assigned as a bending absorption band of O S O in SO4 2− [9,10]. Xiao [13] points out that the animal drug Scorpio contains 28.8–29.2% sulfur and 38.1 g/g natrium. Therefore, we can draw the conclusion that Scorpio contain much sulfate and we can identify the peaks at 1112, 637 and 615 cm−1 are characteristic absorption bands of SO4 2− group, and the peak at 615 cm−1 is assigned to the O S O bending frequency. 3.2. 2D-IR correlation spectra The dynamic spectra of the four drugs were obtained during the thermal perturbation in the range of 50–120 ◦ C. The positions and intensities of characteristic peaks in the dynamic spectra vary very little, which indicates that the state of the sample was not drastically altered by the temperature change. As a typical sample, the dynamic spectra of Cornu Cervi Pantotrichum are shown only (shown in Fig. 4).
The 2D-IR correlation analysis may detect the subtle difference of the response behavior of each group in a molecule under an exterior perturbation, so it can enhances the resolution of spectra and further can obtain much new information which cannot be readily acquired from conventional IR spectra and its derivative spectra. Therefore, there are special advantages using 2D-IR correlation analysis for identification and discrimination of a very complex system. 2D-IR correlation spectra based on thermal perturbation may reveal characteristic behavior of each group in a molecule during the temperature increasing process. Fig. 5 shows the 2D-IR correlation spectra of the four animal drugs studied. Their macro-fingerprint patterns of 2D-IR correlation spectra are obviously different, so we can use them to directly discriminate and identify these drugs. Fig. 5a shows the 2D-IR asynchronous and synchronous correlation spectrum of Cornu Cervi Pantotrichum. Four obvious autopeaks appear at 1562, 1635, 2850 and 2925 cm−1 . Autopeaks at 1635 and 1562 cm−1 are assigned to amide I and II bands, and autopeaks at 2850 and 2925 cm−1 are assigned to CH2 vibration. The presence of autopeaks on synchronous correlation spectrum indicates the thermally induced changes in the intensity of these bands [2]. The positive synchronous cross-peak and negative asynchronous cross-peak at the same coordinate (1635 and 1562 cm−1 ) indicate that the IR intensities of amide I and II bands change in the same direction, but with different rates. Combined with corresponding synchronous peaks, the sign of asynchronous cross-peak can provide additional useful information about
Fig. 4. Dynamic IR spectra of drug Cornu Cervi Pantotrichum during the variable temperature process in the range of 50–120 ◦ C. The marked wavenumbers indicate the position after which the intensity of spectra alter.
Y.-M. Li et al. / Spectrochimica Acta Part A 63 (2006) 565–573
the relative relationship or order of the actual sequence of the process. In Fig. 5a, because the intensity of synchronous cross-peak at 1635 and 1562 cm−1 is positive, while the asynchronous cross-peak at the same coordinate is negative. Thus, we can conclude that the spectral intensity associated with the amide II band at 1562 cm−1 change prior to that of the amide I band at 1635 cm−1 . No cross-peak related to O P O can be observed in Fig. 5a, which indicates there is no obvious change in O P O group. In order to analyze the thermal sensitivity of IR band intensity for O P O group, we expended the 2D-IR correlation spectra in the range of 620–550 cm−1 , shown in Fig. 7a. No autopeaks at 604
569
and 561 cm−1 are observed, which suggests that the O P O group in Ca3 (PO4 )2 contained in Cornu Cervi Pantotrichum is insensitive to the thermal perturbation. Fig. 5b shows the 2D-IR correlation spectra of Cornu Saigae Tataricae. Two strong autopeaks at 1630 and 1562 cm−1 giving rise to amide I and II bands absorption appear, while no autopeak associated to CH2 vibration is observed, which suggests that the CH2 group is less sensitive to the thermal perturbation. The presence of positive synchronous cross-peak at 1630 and 1562 cm−1 and negative asynchronous cross-peak at the same coordinate suggests that the intensities of amide I and II bands
Fig. 5. 2D-IR correlation spectra of four animal drugs (3000–500 cm−1 ): (a) Cornu Cervi Pantotrichum; (b) Cornu Saigae Tataricae; (c) Scorpio; (d) Hirudo. The shaded areas represent negative intensity regions.
570
Y.-M. Li et al. / Spectrochimica Acta Part A 63 (2006) 565–573
Fig. 5. (Continued ).
change in the same direction but with different rates. The cross-peaks at 1630 and 912 cm−1 , 1630 and 953 cm−1 , 1882 and 1562 cm−1 and 1920 and 1562 cm−1 indicate that each pair of vibration groups corresponding to IR bands at 1630 and 912 cm−1 , 1630 and 953 cm−1 , 1882 and 1562 cm−1 , as well as 1920 and 1562 cm−1 are correlated with each other respectively. Because the intensity of synchronous cross-peak at 1630 and 1562 cm−1 is positive, while the asynchronous cross-peak at the same coordinate is negative, the intensity of amide II band at the 1562 cm−1 changes prior to that of the amide I band at 1630 cm−1 .
Fig. 5c shows the 2D-IR correlation spectra of Scorpio. Only a strong autopeak at 1090 cm−1 is observed. The intensity of autopeaks associated with amide I and II bands and CH2 groups is so weak that the corresponding signs cannot be observed. The presence of the negative synchronous cross-peak at 1238 and 1090 cm−1 indicates the IR bands at 1238 and 1090 cm−1 are correlated, and also suggests that the directions of the intensity changes for bands at 1238 and 1090 cm−1 are opposite to each other. The autopeak at 1624 cm−1 and asynchronous cross-peak at 1624 and 1502 cm−1 appear in the expended 2D-IR correlation spectrum of Scorpio in the range of 1700–1500 cm−1 (Fig. 6),
Y.-M. Li et al. / Spectrochimica Acta Part A 63 (2006) 565–573
571
Fig. 6. 2D-IR correlation spectra of Scorpion (1700–1500 cm−1 ).
which further show the autopeak corresponding to amide II band is too weak to be observed. And the autopeak at 615 cm−1 which corresponds to the bending frequency of O S O group is observed from the expended 2D-IR correlation spectra of the drug Scorpio in the range of 650–600 cm−1 (Fig. 7b). The result indicates that the intensities associated with O S O group change strongly, but IR bands giving rise to O S O and organic groups are not correlated since no corresponding cross-peaks can be observed in Fig. 5c. Fig. 5d shows the 2D-IR correlation spectra of Hirudo. The presence of autopeaks at 1635,1562 cm−1 associated with amide I and II bands and no autopeaks related to CH2 groups indicate that amide I and II bands are sensitive to the thermal perturbation, while the CH2 groups is less sensitive. The presence of the synchronous and asynchronous peaks at coordinates (1635 and 1562 cm−1 ), (2852 and 1562 cm−1 ) and (2922 and 1562 cm−1 ) suggest that the IR bands at 1635 and 1562 cm−1 , 2852 and 1562 cm−1 , as well as 2922 and 1562 cm−1 are correlated with each other. It means that the amide group and the CH2 group experienced by thermally induced changes may be connected. The positive intensity of the three cross-peaks indicates that the their intensities of amide I, II and CH2 group change in the same direction, while the presence of asynchronous peaks at the same coordinates suggests that the intensity changes occur at different rates. Because the correlation intensity of synchronous cross-peak at 1635 and 1562 cm−1 is positive, while the asynchronous cross-peak at the same coordinate is negative, we can again conclude that the intensity of the amide II band at 1562 cm−1 changes prior to that of the amide I band at 1635 cm−1 . Signs of synchronous and asynchronous cross-peaks at 2852 and 1635 cm−1 are both positive. Therefore, the intensity of band associated with the CH2 group changes prior to that of the amide I band. The sign of synchronous cross-peak at 2852 and 1562 cm−1 is positive, while
that of the asynchronous cross-peak at the same coordinate is negative, so the intensity of the amide II band changes prior to that of the CH2 group. As we noted, the onset of the spectral intensity changes of amide II and amide I bands during the heating is different. On one hand, there is a 43 conjugative bond of the O C N group in amide RCONRR . When the temperature rises, the conjugative bond turns out to be unstable. The C N bond in the 43 conjugative system converts from double bond to single bond, but the C O bond remains double bond. The R and R groups which are connected with the C N bond turn out to rotate relatively freely in solution. So the intensity change of the amide II band associated mainly with the stretching frequency of C N bond occurs prior to that of the amide I band associated with the stretching frequency of C O band. On the other hand, amide II vibrations, which may involve much more of the hydrogen bonded local structures of amide groups in the system compared to the C O dominated amide I vibration, may undergo thermally induced changes at a temperature much lower than the other mode. Thus, we may be observing the differential susceptibility of the thermally induced local structural changes. This sequence is consistent with that deduced from the 2D-IR correlation spectra. In sum, the response of the organic components of the drugs to the thermal perturbation is much stronger than that of the inorganic components. The intensity changes of amide I and II bands in the animal drugs Cornu Cervi Pantotrichum, Cornu Saigae Tataricae and Hirudo are apparent and in the same directions, but with different rates. Moreover, the intensity change of amide II band occurs prior to that of amide I band. However, no intensity change in amide II band in drug Scorpio is observed. The only visible autopeak corresponding to the band intensity changes of the O S O group shows no strong correlation with organic compounds. The O P O group shows no band intensity change.
572
Y.-M. Li et al. / Spectrochimica Acta Part A 63 (2006) 565–573
Fig. 7. 2D-IR correlation spectra of: (a) Cornu Cervi Pantotrichum (620–550 cm−1 ) and (b) Scorpion (650–600 cm−1 ). The shaded areas represent negative intensity regions.
4. Conclusions IR spectra and 2D-IR correlation spectra of the four traditional Chinese animal drugs have exclusive macro-fingerprint characteristics of their own, which can be effectively used to identify and discriminate one drug from the other. The broad peak at 1033 cm−1 and peaks at 604 and 561 cm−1 can be used to identify the drug Cornu Cervi Pantotrichum, and the sharp peak at 1112 cm−1 and peaks at 637 and 615 cm−1 can be used to identify the drug Scorpio. For drugs of Cornu Cervi Pantotrichum, Cornu Saigae Tataricae and Hirudo, the intensities of amide I and II bands change in the same directions but with different rates. The intensity of amide II band changes prior to that of amide I band not only because amide II vibrations may involve much more of the hydrogen bonded
local structures of amide groups in the system compared to the C O dominated amide I vibration, but also because the C N bond in the 43 conjugative system converts from double bond to single bond during the heating process. Thus, the 2D-IR correlation spectroscopy combined with IR spectroscopy can be used to effectively identify the animal drugs. 2D-IR correlation analysis of animal drugs provides further valuable information for pharmacological research of the traditional Chinese animal drugs.
Acknowledgements This work is sponsored by the State Administration of Traditional Chinese Medicine of the People’s Republic of
Y.-M. Li et al. / Spectrochimica Acta Part A 63 (2006) 565–573
China (2001ZDZX01) and grand support of Ministry of Science and Technology of the People’s Republic of China (2002BA906A29-4). References [1] [2] [3] [4] [5]
I. Noda, J. Am. Chem. Soc. 111 (21) (1989) 8116–8118. I. Noda, Appl. Spectrosc. 47 (1993) 1329–1336. I. Noda, Appl. Spectrosc. 44 (4) (1990) 550–561. Y. Ozaki, Y. Liu, I. Noda, Appl. Spectrosc. 51 (1997) 526. (a) S.Q. Sun, Q. Zhou, Z. Qin, J.X. Tao, Chinese Patent ZL 03123389.9; (b) S.Q. Sun, Z. Qin, Q. Zhou, J.X. Tao, Chinese Patent ZL 03122935.2; (c) S.Q. Sun, Q. Zhou, Z. Qin, Chinese Patent ZL 03104765.3.
573
[6] L. Zuo, S.Q. Sun, Q. Zhou, J.X. Tao, I. Noda, J. Pharm. Biomed. Anal. 30 (2003) 1491–1498. [8] S.Q. Sun, Q. Zhou, Z. Qin, Atlas of Two-Dimensional Correlation Infrared Spectroscopy of Chinese Medicine, Chemical Industry Press, Beijing, 2003 (in Chinese). [9] R. Nyquist, R. Kagel, Infrared Spectra of Inorganic Compounds, Academic Press, New York/London, 1971, pp. 288–289. [10] K. Nakamoto, Infrared and Raman Spectra of Inorganic and Coordination Compounds, Wiley, New York, 1986, pp. 130–138, 477. [11] J. Zhen, L. Zhang, H. Zhou, Chin. J. Shenyang Pharm. Univ. 13 (3) (1996) 196–200. [12] The Infrared Spectroscopy Committee of Chicago Society for Coatings Technology, An Infrared Spectroscopy Atlas for the Coatings Industry, Federation of Societies for Coatings Technology, Philadelphia, PA, 1980, p. 533. [13] P.G. Xiao, Modern Chinese Materia Medica, Chemistry Industry Press, 2002, p. 76.
Study of traditional Chinese animal drugs using FT-IR and 2D-IR correlation spectroscopy Ying-Ming Li a , Su-Qin Sun a,∗ , Qun Zhou a , Jia-Xun Tao a , Isao Noda b a
Key Laboratory of Bioorganic Phosphorus Chemistry & Chemical Biology (Ministry of Education), Department of Chemistry, Tsinghua University, Beijing 100084, PR China b The Procter & Gamble Co., West Chester, OH 45069, USA Received 28 March 2005; received in revised form 2 June 2005; accepted 3 June 2005
Abstract Using FT-IR and 2D-IR correlation spectroscopy, the intensity changes and their sequence of amide I and II bands of four traditional Chinese animal drugs (Cornu Cervi Pantotrichum, Cornu Saigae Tataricae, Scorpio and Hirud) under thermal perturbation are studied, and component of Ca3 (PO4 )2 in the drug of Cornu Cervi Pantotrichum and sulfates components in the drug of scorpion are identified. The drug of Cornu Cervi Pantotrichum contains inorganic salt Ca3 (PO4 )2 and the drug of Scorpio contains sulfates. It is assigned that the bands of 604 and 561 cm−1 belong to the component of Ca3 (PO4 )2 , and the bands of 637 and 615 cm−1 belong to sulfates. Organic components of these drugs respond to the thermal perturbation far stronger than that of the inorganic components. The intensities of amide I and II bands in the drugs, except for amide II band in drug Scorpio, change strongly. For the drugs of Cornu Cervi Pantotrichum, Cornu 4 Saigae Tataricae and Hirudo, the intensity changes of amide II band occurs prior to that of amide I band. The C N bond in the conjugative system converts from double bond to single bond, but the C O bond remains double bond during the heating process. 3 On the other hand, amide II vibrations, which may involve much more of the hydrogen bonded local structures of amide groups in the system compared to the C O dominated amide I vibration, may undergo thermally induced changes at a temperature much lower than the other mode. The traditional Chinese animal drugs can be identified rapidly and non-separately by using FT-IR and 2D-IR correlation spectroscopy. © 2005 Elsevier B.V. All rights reserved. Keywords: FT-IR; 2D-IR correlation spectroscopy; Animal drugs
1. Introduction Perturbation-based 2D correlation spectroscopy was originally developed in IR field. Compared with conventional spectra, 2D correlation spectra can enhance the spectral resolution by spreading peaks over the second dimension, and simplify the complex spectra consisting of many overlapped peaks, as well as identify various inter and intra molecular interactions through selective correlation of IR peaks [1–3]. The 2D correlation technique has been widely used in traditional spectroscopy, such as IR, Raman, UV and fluorescence. As for 2D-IR correlation spectroscopy, there ∗
Corresponding author. Tel.: +86 10 62787661; fax: +86 10 62770327. E-mail address: [email protected] (S.-Q. Sun).
1386-1425/$ – see front matter © 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.saa.2005.06.004
have been studies on many aspects, including biologically important systems like N-methylacetamide (NMA) [4]. As a non-separate, non-extraction, rapid and convenient method, 2D-IR spectroscopy combined with IR spectroscopy which we called “IR macro-fingerprint method” has extended the application of 2D correlation spectroscopy from simple compounds to very complex systems, such as the field of herbal medicines, including raw herbs [5a], processed medicines, formula granules [5b] and injections [5c]. The IR macro-fingerprint method follows the integrate principle of traditional Chinese medicine, and it does not lose original natural instinct and compatibility of traditional Chinese medicine. This method has been successfully used to the trace and analysis of the deteriorative process of injections, such as Qingkailing injection [6]. It proves that the dete-
Y.-M. Li et al. / Spectrochimica Acta Part A 63 (2006) 565–573
566
rioration of the Qingkailing injection in air is due to the oxidation of flavones and/or the variations in conformation and structure of the flavones compounds. This method also has been applied to the quality control of traditional Chinese medicines. We have made over 200 2D-IR spectra of traditional Chinese medicines and further proposed multistep IR identification method to identify herbal medicines: the IR spectra as the first step identification; the second derivative spectra as the second step identification; the 2D-IR correlation spectra as the third step identification [8]. It presents a new idea for the analysis of complex systems. The so-called animal drugs have reliable and special curative effects and are the important part of traditional Chinese medicine. Their main components contain proteins. The IR spectroscopy is rarely applied directly to the study of animal drugs. In this paper, based on the above method, we emphasized and analyzed the macro-fingerprint features of animal drugs Cornu Cervi Pantotrichum, Cornu Saigae Tataricae, Scorpio and Hirud, as well as their response to thermal perturbations, which is very important to the identification of animal drugs and further to realize their pharmacological efficacies.
The reagents of Ca3 (PO4 )2 and Na2 SO4 are used as received. All animal drugs were identified and provided by the National Institute for the Control of Pharmaceutical and Biological Products of China. 2.2. Procedure Each sample of animal drugs was ground into powder with over 200 meshes and then blended with KBr powder, ground again and pressed into a tablet. After that their IR spectra were collected at room temperature. The dynamic spectra were collected at different temperatures from 50 to 120 ◦ C at interval of 10 ◦ C and were processed by baseline correction. We made baseline autocorrection of spectra during the temperature increasing process before 2D correlation analysis to overcome the baseline drifts. 2D-IR correlation spectra were obtained by the treatment of a series of dynamic spectra with 2D-IR correlation analysis software programmed by our group.
3. Results and discussion 2. Experiment
3.1. IR spectra
2.1. Apparatus and samples
Fig. 1 shows the IR spectra of the four animal drugs. The assignments of characteristic peaks of the spectra are shown in Table 1. There are some similarities of the spectra, such as the strong absorption peaks of the amide bands, since the animal drugs contain a high level of protein components. The amide I bands are assigned to the stretching frequency of C O groups, and the amide II bands are mainly associated
Spectrum GX FT-IR spectrometer (Perkin-Elmer) is equipped with a DTGS detector and portable programmable temperature controller (Model 50–886, Love Control Corporation). IR spectra were recorded from an accumulation of 32 scans in 4000–400 cm−1 range with a resolution of 4 cm−1 .
Fig. 1. The IR spectra of four animal drugs: (a) Cornu Cervi Pantotrichum; (b) Cornu Saigae Tataricae; (c) Scorpion; (d) Hirudo.
Y.-M. Li et al. / Spectrochimica Acta Part A 63 (2006) 565–573
567
Table 1 The assignments of main IR absorption bands of the drugs Animal drug
Amide I (cm−1 )
Amide II (cm−1 )
δ(CH2 , CH3 ) (cm−1 )
γ(CH2 , CH3 ) (cm−1 )
δO–P–O (cm−1 )
δO–S–O (cm−1 )
Cornu Cervi Pantotrichum Cornu Saigae Tataricae Scorpio Hirudo
1659 1657 1662 1659
1546 1517 1531 1537
2956, 2925 2960, 2933 2959, 2927 2959, 2927
2853, 1453 2875, 1451 2855, 1442 2855, 1453
561 – – –
– – 615 –
Fig. 2. Comparison of IR spectra (left) and second derivative spectra (right) of: (a) Cornu Cervi Pantotrichum and (b) Ca3 (PO4 )2 .
with stretching frequency of C N bond, as well as bending vibration of N H groups. However, the positions of these absorption bands related with the amide groups are different for different animal drugs. For example, the characteristic amide bands appear at 1659 and 1546 cm−1 for Cornu Cervi Pantotrichum; at 1657 and 1517 cm−1 for Cornu Saigao; at 1662 and 1531 cm−1 for Scorpion; at 1659 and 1537 cm−1 for Hirudo. Some unique peaks can be observed such as peaks at 1033, 604 and 561 cm−1 for Cornu Cervi Pantotrichum, shown in Fig. 1a and peaks at 1112, 637 and 615 cm−1 for Scorpion, shown in Fig. 1c. We measured the IR spectra and the second derivative spectra of Ca3 (PO4 )2 and Na2 SO4 , and compared these spectra with the spectra of nearly fifty kinds of inorganic compounds including Mg3 (PO4 )2 , Na3 PO4 , etc., and found that in the range of 1100–400 cm−1 the spectra of Cornu Cervi Pantotrichum is very similar to that of Ca3 (PO4 )2 : the peaks at 1033, 961, 604 and 561 cm−1 in IR spectrum of Cornu Cervi Pantotrichum correspond to
the absorption bands at 1031, 962, 603 and 564 cm−1 of Ca3 (PO4 )2 and the peaks at 1033, 961, 604 and 561 cm−1 in the second derivative spectrum correspond to the absorption bands at 1030, 961, 602 and 563 cm−1 of Ca3 (PO4 )2 , shown in Fig. 2, and the spectra of Scorpio is very similar to that of sulfate in the range of 1200–400 cm−1 : the peaks at 1112, 637 and 615 cm−1 in the IR spectrum of Scorpio correspond to the peaks at 1118, 638 and 618 cm−1 of Na2 SO4 and the peaks at 638 and 614 cm−1 in the second derivative spectra of Scorpio correspond to the absorption bands at 639 and 618 cm−1 of Na2 SO4 , shown in Fig. 3. In general, the absorption bands of PO4 3− appear in ranges within 1120–940 cm−1 and within 650–540 cm−1 , and the peak at 567 cm−1 is assigned to bending absorption band of O P O for PO4 3− [9,10]. Zhen et al. [11] considered that in the spectrum of the drug Cornu Cervi Pantotrichum, the peaks below 1100 cm−1 is related to the C C stretching frequencies with medium intensities and the O P O vibrations
Fig. 3. Comparison of IR spectra (left) and second derivative spectra (right) of: (a) Scorpio and (b) Na2 SO4 .
568
Y.-M. Li et al. / Spectrochimica Acta Part A 63 (2006) 565–573
of the calcium phosphate components. The peaks in the range of 1100–400 cm−1 in Fig. 2a are in agreement with the characteristic absorption bands of bone black, which contains 90–94% calcium phosphate and 6–10% carbon [12]. So, it is concluded that the animal drug Cornu Cervi Pantotrichum contain much Ca3 (PO4 )2 components, of which the peaks at 604 and 561 cm−1 (the O P O bending vibration) are characteristic absorption bands. The bending vibrations of O S O in SO4 2− often appear at range within 960–1030 cm−1 with strong and sharp peak and at range within 570–680 cm−1 with medium intensity. And the peak at 615 cm−1 has been assigned as a bending absorption band of O S O in SO4 2− [9,10]. Xiao [13] points out that the animal drug Scorpio contains 28.8–29.2% sulfur and 38.1 g/g natrium. Therefore, we can draw the conclusion that Scorpio contain much sulfate and we can identify the peaks at 1112, 637 and 615 cm−1 are characteristic absorption bands of SO4 2− group, and the peak at 615 cm−1 is assigned to the O S O bending frequency. 3.2. 2D-IR correlation spectra The dynamic spectra of the four drugs were obtained during the thermal perturbation in the range of 50–120 ◦ C. The positions and intensities of characteristic peaks in the dynamic spectra vary very little, which indicates that the state of the sample was not drastically altered by the temperature change. As a typical sample, the dynamic spectra of Cornu Cervi Pantotrichum are shown only (shown in Fig. 4).
The 2D-IR correlation analysis may detect the subtle difference of the response behavior of each group in a molecule under an exterior perturbation, so it can enhances the resolution of spectra and further can obtain much new information which cannot be readily acquired from conventional IR spectra and its derivative spectra. Therefore, there are special advantages using 2D-IR correlation analysis for identification and discrimination of a very complex system. 2D-IR correlation spectra based on thermal perturbation may reveal characteristic behavior of each group in a molecule during the temperature increasing process. Fig. 5 shows the 2D-IR correlation spectra of the four animal drugs studied. Their macro-fingerprint patterns of 2D-IR correlation spectra are obviously different, so we can use them to directly discriminate and identify these drugs. Fig. 5a shows the 2D-IR asynchronous and synchronous correlation spectrum of Cornu Cervi Pantotrichum. Four obvious autopeaks appear at 1562, 1635, 2850 and 2925 cm−1 . Autopeaks at 1635 and 1562 cm−1 are assigned to amide I and II bands, and autopeaks at 2850 and 2925 cm−1 are assigned to CH2 vibration. The presence of autopeaks on synchronous correlation spectrum indicates the thermally induced changes in the intensity of these bands [2]. The positive synchronous cross-peak and negative asynchronous cross-peak at the same coordinate (1635 and 1562 cm−1 ) indicate that the IR intensities of amide I and II bands change in the same direction, but with different rates. Combined with corresponding synchronous peaks, the sign of asynchronous cross-peak can provide additional useful information about
Fig. 4. Dynamic IR spectra of drug Cornu Cervi Pantotrichum during the variable temperature process in the range of 50–120 ◦ C. The marked wavenumbers indicate the position after which the intensity of spectra alter.
Y.-M. Li et al. / Spectrochimica Acta Part A 63 (2006) 565–573
the relative relationship or order of the actual sequence of the process. In Fig. 5a, because the intensity of synchronous cross-peak at 1635 and 1562 cm−1 is positive, while the asynchronous cross-peak at the same coordinate is negative. Thus, we can conclude that the spectral intensity associated with the amide II band at 1562 cm−1 change prior to that of the amide I band at 1635 cm−1 . No cross-peak related to O P O can be observed in Fig. 5a, which indicates there is no obvious change in O P O group. In order to analyze the thermal sensitivity of IR band intensity for O P O group, we expended the 2D-IR correlation spectra in the range of 620–550 cm−1 , shown in Fig. 7a. No autopeaks at 604
569
and 561 cm−1 are observed, which suggests that the O P O group in Ca3 (PO4 )2 contained in Cornu Cervi Pantotrichum is insensitive to the thermal perturbation. Fig. 5b shows the 2D-IR correlation spectra of Cornu Saigae Tataricae. Two strong autopeaks at 1630 and 1562 cm−1 giving rise to amide I and II bands absorption appear, while no autopeak associated to CH2 vibration is observed, which suggests that the CH2 group is less sensitive to the thermal perturbation. The presence of positive synchronous cross-peak at 1630 and 1562 cm−1 and negative asynchronous cross-peak at the same coordinate suggests that the intensities of amide I and II bands
Fig. 5. 2D-IR correlation spectra of four animal drugs (3000–500 cm−1 ): (a) Cornu Cervi Pantotrichum; (b) Cornu Saigae Tataricae; (c) Scorpio; (d) Hirudo. The shaded areas represent negative intensity regions.
570
Y.-M. Li et al. / Spectrochimica Acta Part A 63 (2006) 565–573
Fig. 5. (Continued ).
change in the same direction but with different rates. The cross-peaks at 1630 and 912 cm−1 , 1630 and 953 cm−1 , 1882 and 1562 cm−1 and 1920 and 1562 cm−1 indicate that each pair of vibration groups corresponding to IR bands at 1630 and 912 cm−1 , 1630 and 953 cm−1 , 1882 and 1562 cm−1 , as well as 1920 and 1562 cm−1 are correlated with each other respectively. Because the intensity of synchronous cross-peak at 1630 and 1562 cm−1 is positive, while the asynchronous cross-peak at the same coordinate is negative, the intensity of amide II band at the 1562 cm−1 changes prior to that of the amide I band at 1630 cm−1 .
Fig. 5c shows the 2D-IR correlation spectra of Scorpio. Only a strong autopeak at 1090 cm−1 is observed. The intensity of autopeaks associated with amide I and II bands and CH2 groups is so weak that the corresponding signs cannot be observed. The presence of the negative synchronous cross-peak at 1238 and 1090 cm−1 indicates the IR bands at 1238 and 1090 cm−1 are correlated, and also suggests that the directions of the intensity changes for bands at 1238 and 1090 cm−1 are opposite to each other. The autopeak at 1624 cm−1 and asynchronous cross-peak at 1624 and 1502 cm−1 appear in the expended 2D-IR correlation spectrum of Scorpio in the range of 1700–1500 cm−1 (Fig. 6),
Y.-M. Li et al. / Spectrochimica Acta Part A 63 (2006) 565–573
571
Fig. 6. 2D-IR correlation spectra of Scorpion (1700–1500 cm−1 ).
which further show the autopeak corresponding to amide II band is too weak to be observed. And the autopeak at 615 cm−1 which corresponds to the bending frequency of O S O group is observed from the expended 2D-IR correlation spectra of the drug Scorpio in the range of 650–600 cm−1 (Fig. 7b). The result indicates that the intensities associated with O S O group change strongly, but IR bands giving rise to O S O and organic groups are not correlated since no corresponding cross-peaks can be observed in Fig. 5c. Fig. 5d shows the 2D-IR correlation spectra of Hirudo. The presence of autopeaks at 1635,1562 cm−1 associated with amide I and II bands and no autopeaks related to CH2 groups indicate that amide I and II bands are sensitive to the thermal perturbation, while the CH2 groups is less sensitive. The presence of the synchronous and asynchronous peaks at coordinates (1635 and 1562 cm−1 ), (2852 and 1562 cm−1 ) and (2922 and 1562 cm−1 ) suggest that the IR bands at 1635 and 1562 cm−1 , 2852 and 1562 cm−1 , as well as 2922 and 1562 cm−1 are correlated with each other. It means that the amide group and the CH2 group experienced by thermally induced changes may be connected. The positive intensity of the three cross-peaks indicates that the their intensities of amide I, II and CH2 group change in the same direction, while the presence of asynchronous peaks at the same coordinates suggests that the intensity changes occur at different rates. Because the correlation intensity of synchronous cross-peak at 1635 and 1562 cm−1 is positive, while the asynchronous cross-peak at the same coordinate is negative, we can again conclude that the intensity of the amide II band at 1562 cm−1 changes prior to that of the amide I band at 1635 cm−1 . Signs of synchronous and asynchronous cross-peaks at 2852 and 1635 cm−1 are both positive. Therefore, the intensity of band associated with the CH2 group changes prior to that of the amide I band. The sign of synchronous cross-peak at 2852 and 1562 cm−1 is positive, while
that of the asynchronous cross-peak at the same coordinate is negative, so the intensity of the amide II band changes prior to that of the CH2 group. As we noted, the onset of the spectral intensity changes of amide II and amide I bands during the heating is different. On one hand, there is a 43 conjugative bond of the O C N group in amide RCONRR . When the temperature rises, the conjugative bond turns out to be unstable. The C N bond in the 43 conjugative system converts from double bond to single bond, but the C O bond remains double bond. The R and R groups which are connected with the C N bond turn out to rotate relatively freely in solution. So the intensity change of the amide II band associated mainly with the stretching frequency of C N bond occurs prior to that of the amide I band associated with the stretching frequency of C O band. On the other hand, amide II vibrations, which may involve much more of the hydrogen bonded local structures of amide groups in the system compared to the C O dominated amide I vibration, may undergo thermally induced changes at a temperature much lower than the other mode. Thus, we may be observing the differential susceptibility of the thermally induced local structural changes. This sequence is consistent with that deduced from the 2D-IR correlation spectra. In sum, the response of the organic components of the drugs to the thermal perturbation is much stronger than that of the inorganic components. The intensity changes of amide I and II bands in the animal drugs Cornu Cervi Pantotrichum, Cornu Saigae Tataricae and Hirudo are apparent and in the same directions, but with different rates. Moreover, the intensity change of amide II band occurs prior to that of amide I band. However, no intensity change in amide II band in drug Scorpio is observed. The only visible autopeak corresponding to the band intensity changes of the O S O group shows no strong correlation with organic compounds. The O P O group shows no band intensity change.
572
Y.-M. Li et al. / Spectrochimica Acta Part A 63 (2006) 565–573
Fig. 7. 2D-IR correlation spectra of: (a) Cornu Cervi Pantotrichum (620–550 cm−1 ) and (b) Scorpion (650–600 cm−1 ). The shaded areas represent negative intensity regions.
4. Conclusions IR spectra and 2D-IR correlation spectra of the four traditional Chinese animal drugs have exclusive macro-fingerprint characteristics of their own, which can be effectively used to identify and discriminate one drug from the other. The broad peak at 1033 cm−1 and peaks at 604 and 561 cm−1 can be used to identify the drug Cornu Cervi Pantotrichum, and the sharp peak at 1112 cm−1 and peaks at 637 and 615 cm−1 can be used to identify the drug Scorpio. For drugs of Cornu Cervi Pantotrichum, Cornu Saigae Tataricae and Hirudo, the intensities of amide I and II bands change in the same directions but with different rates. The intensity of amide II band changes prior to that of amide I band not only because amide II vibrations may involve much more of the hydrogen bonded
local structures of amide groups in the system compared to the C O dominated amide I vibration, but also because the C N bond in the 43 conjugative system converts from double bond to single bond during the heating process. Thus, the 2D-IR correlation spectroscopy combined with IR spectroscopy can be used to effectively identify the animal drugs. 2D-IR correlation analysis of animal drugs provides further valuable information for pharmacological research of the traditional Chinese animal drugs.
Acknowledgements This work is sponsored by the State Administration of Traditional Chinese Medicine of the People’s Republic of
Y.-M. Li et al. / Spectrochimica Acta Part A 63 (2006) 565–573
China (2001ZDZX01) and grand support of Ministry of Science and Technology of the People’s Republic of China (2002BA906A29-4). References [1] [2] [3] [4] [5]
I. Noda, J. Am. Chem. Soc. 111 (21) (1989) 8116–8118. I. Noda, Appl. Spectrosc. 47 (1993) 1329–1336. I. Noda, Appl. Spectrosc. 44 (4) (1990) 550–561. Y. Ozaki, Y. Liu, I. Noda, Appl. Spectrosc. 51 (1997) 526. (a) S.Q. Sun, Q. Zhou, Z. Qin, J.X. Tao, Chinese Patent ZL 03123389.9; (b) S.Q. Sun, Z. Qin, Q. Zhou, J.X. Tao, Chinese Patent ZL 03122935.2; (c) S.Q. Sun, Q. Zhou, Z. Qin, Chinese Patent ZL 03104765.3.
573
[6] L. Zuo, S.Q. Sun, Q. Zhou, J.X. Tao, I. Noda, J. Pharm. Biomed. Anal. 30 (2003) 1491–1498. [8] S.Q. Sun, Q. Zhou, Z. Qin, Atlas of Two-Dimensional Correlation Infrared Spectroscopy of Chinese Medicine, Chemical Industry Press, Beijing, 2003 (in Chinese). [9] R. Nyquist, R. Kagel, Infrared Spectra of Inorganic Compounds, Academic Press, New York/London, 1971, pp. 288–289. [10] K. Nakamoto, Infrared and Raman Spectra of Inorganic and Coordination Compounds, Wiley, New York, 1986, pp. 130–138, 477. [11] J. Zhen, L. Zhang, H. Zhou, Chin. J. Shenyang Pharm. Univ. 13 (3) (1996) 196–200. [12] The Infrared Spectroscopy Committee of Chicago Society for Coatings Technology, An Infrared Spectroscopy Atlas for the Coatings Industry, Federation of Societies for Coatings Technology, Philadelphia, PA, 1980, p. 533. [13] P.G. Xiao, Modern Chinese Materia Medica, Chemistry Industry Press, 2002, p. 76.