TY - JOUR
AU1 - Li,, Yue
AU2 - Duan,, Jinlian
AU3 - Xia,, Heng
AU4 - Shu,, Bin
AU5 - Duan,, Weigang
AB - Abstract Macromolecular substances in traditional Chinese medicine injections (TCMIs) are expected to be a main dangerous factor causing anaphylactic or anaphylactoid reaction. The main aim of the study was to verify the macromolecular substances’ anaphylactic or anaphylactoid reaction in guinea pigs and establish a size-exclusive chromatographic method to detect them. The macromolecular substances from six TCMIs (Danshen injection, Dengzhanxixin injection, Honghua injection, Qingkailing injection, Shuanghuanglian injection and Shuxuening injection) were prepared by removing substances with molecular weight less than 10 kDa with an ultra-filter. The anaphylactic and anaphylactoid reactions caused by original TCMIs, injections rich in or free of macromolecules were assayed in guinea pigs. The relationship between the amount of the macromolecular substances and peak area of chromatogram was established by size-exclusive chromatography. Injections free of macromolecules were not likely to cause anaphylactic and anaphylactoid reactions, but injections rich in macromolecular substances were more likely to do so. If the macromolecular substances with molecular weight bigger than 10 kDa were removed, the signal of macromolecular substances in TCMIs was quantitatively reduced. All the results suggested that macromolecular substances in TCMIs are a dangerous factor causing safety problems, and the macromolecular substances can be quantitatively detected with size-exclusive chromatography. traditional Chinese medicine injection, macromolecular substances, anaphylactoid reaction, anaphylactic reaction, size-exclusive chromatography Introduction Traditional Chinese medicine injection (TCMI) is a unique dosage form in Chinese Mainland. Classic TCMI was made from one or more raw natural drug(s), mainly medicinal plants, based on the theory of traditional Chinese medicine (TCM). Chaihu injection, also named ‘Baobolier’, was the first TCMI manufactured as early as 1940s [1]. Since then, more than one thousand TCMIs were ever developed and approved by Chinese authorities. However, most of them were abandoned or suspended recently because of their safety problems, and there are only 132 TCMIs left approved for clinical practice [2]. Of course, TCMIs made a big contribution for Chinese health [3], though their problems should be solved. According to chemicals and materials they contained, TCMIs can be grouped into monomeric injections (containing only one active compound), active fraction injections (containing several active compounds shared similar structures), single raw drug injections (an extraction from one raw natural drug) and multiple raw drug injections (an extraction from two or more raw natural drugs). Among them, the first two were regarded as modern TCMIs and the latter two were classic ones. In fact, most TCMIs contained two or more active compounds and impurities, and the complex mixture, especially from the classic TCMIs, become a big challenge for modern scientists to analyze and evaluate. TCMIs’ safety problems become a tough clinic issue recently. Injections with frequent adverse reports included Danshen injection (DS) [4], Dengzhanxixin injection (DZXX) [5], Honghua injection (HH) [6], Qingkailing injection (QKL) [7], Shuanghuanglian injection (SHL) [8] and Shuxuening injection (SXN) [9], all of which were classic TCMIs widely used in the clinic. All the aforementioned TCMIs are approved for intravenous injection. Their safety problems include arrhythmia, transient hypotension, asthmatic attack, respiratory depression, liver dysfunction, dermatitis, and so on [4–12]. Moreover, shock or death events were frequently reported [11]. Most safety problems appeared in 24 h, especially 30 min after administration [11, 12]. According to the recent analysis, most safety problems caused by TCMIs were believed belong to anaphylactic reaction (allergy) [13, 14] or more frequently anaphylactoid reaction [7]. Unlike TCMIs, the oral TCMs with similar formula seldom caused safety problems, and if happened, much weaker [12]. Therefore, it can be deduced that substances in TCMIs that can be kept out by the gastrointestinal tract could be the main factor causing the problems. The macromolecular substances are the important members nonabsorbable [15, 16]. TCMIs are extracts from plants and/or animals, containing active components and impurities. As impurities, macromolecular substances were proved to exist in TCMIs [17], and believed to be the main reason causing anaphylactic or anaphylactoid reaction. The macromolecular substances in TCMIs include proteins, tannins and resins. The methods how to independently detect them have been developed and documented in Chinese Pharmacopoeia (CP) since 2000 edition and were improved recently [18–20]. However, the methods are not sensitive enough to detect the macromolecular substances in TCMIs, and the total macromolecular substances in them were never objectively evaluated. Though there was a limited report that detected the macromolecular substances in a TCMI with HPLC [21], most of them were not successfully detected by chromatography. In order to solve the safety problems, the main problems caused by macromolecular substances are needed to be verified, and a sensitive method to detect them is needed to be established. Materials and Methods Materials Six TCMIs including DS (Batch No.: 1005104), DZXX (Batch No.: 20130533), HH (Batch No.: 1031504213), QKL (Batch No.: 1003272), SHL (Batch No.: 2010032437) and SXN (Batch No.: 1041611272) were provided by manufacturers or obtained from markets. All of them were approved for clinical application in Chinese Mainland. Among them, DS was made from the total extraction of the root or rhizome of Salvia miltiorrhiza Bunge, DZXX was from the total extraction of the whole plant of Erigeron breviscapus (Vaniot) Hand.-Mazz, HH was from the total extraction of the flower of Carthamus tinctorius L., QKL was from the total extraction of a compound formula containing cholic acid, mother-of-pearl, deoxycholic acid, fruit of Gardenia jasminoides Ellis, buffalo horn, root of Isatis tinctoria L., baicalin, and flower of Lonicera japonica Thunb., SHL was from the total extraction of a formula containing flower of L. japonica Thunb., root of Scutellaria baicalensis Georgi, and fruit of Forsythia suspensa (Thunb.) Vahl, and SXN was made of flavonoids from ginkgo leaves. All the injections were approved for intravenous injection. Guinea pigs weighing 250–350 g were obtained from Kunming Medical University, Kunming, China [Certification No. SCXK (Dian) K2015–0005]. The animals were maintained at 22°C, with a humidity of 45–55% under natural light and with free approach to food and water. Sulfosalicylic acid (Lot: A610610) and ammonium sulfate (Lot: A501076) were purchased from Sangon Biotech (Shanghai) Co., Ltd. (Shanghai, China). Egg white solution of 1% was self-made by diluting egg white with normal saline for injection and passed through a 0.22-μm pore size filter. The HPLC system of LC-10Tvp was manufactured by Surwit Technology Inc. (Hangzhou, China). The column for size-exclusive chromatography was Sephadex G-10 column (10 mm × 300 mm, 40–120 μm), manufactured by Tianjin Puxiang Technology Ltd. (Tianjin, China). The multiple micro-plate reader of Infinite 200 pro was manufactured by Tecan Group (Männedorf, Switzerland). The lyophilizer of FDU-1100 was manufactured by EYELA (Tokyo, Japan). Ultrapure water was obtained from a Milli-Q water purification system (EMD Millipore Group, Darmstadt, Germany). Ultra-filters of 10 kDa (UFC901096, molecular sieves) were manufactured by EMD Millipore Group (Cork, Ireland). Other instruments or reagents used in the present study were made in China. Preparation of TCMI Free of or Rich in Macromolecular Substances and TCMI Macromolecular Substance The original TCMIs are brown or alike, and 20 ml of them were gradually added to the upper bowl of the ultra-filter. The ultra-filter was spun at 3,500 rpm and at 4°C until nine-tenth solution was filtered through. The ultra-filtered solution ‘free’ of macromolecular substances was collected and kept at −40°C or used as ‘injection free of macromolecule (TCMI-F)’ immediately. The concentrated solution in the upper bowl was used as ‘injection rich in macromolecules (TCMI-R)’. A portion of TCMI-R was further rinsed with ultrapure water of same volume six times and spun under the same conditions until the filtered solution was colorless in order to obtain macromolecular reference. The concentrated solution containing macromolecular substances (>10 kDa) and almost free of ‘small molecular’ substances (<10 kDa) in the upper bowl was collected and lyophilized. The lyophilized powder was weighted and used as the macromolecular reference. Solid content in the TCMIs was determined with lyophilization. Briefly, TCMI of 10 ml in the glass ampoule was transferred to a plastic tube of 50 ml and frozen at −40°C overnight. Then, the frozen TCMI was put in the lyophilizer and lyophilized for more than 48 h. After that, the tube was weighted every 3 h, and when the total weight of the tube decreased less than 0.5 mg, the lyophilization was ended. The net solid weight was calculated by subtracting the tube weight from the total weight. Anaphylactic and Anaphylactoid Reactions Detected in Guinea Pig Every six guinea pigs including three males and three females (without pregnancy) were randomly arranged in a group to detect anaphylactic or anaphylactoid reactions. The animal experiment was approved by the Animal Care and Use Committee of Jiangsu Provicial Institute of Materia Medica (Approved No. LL-20170830-02), Nanjing Tech University, Nanjing China. The method to detect anaphylactic reaction was performed according to CP of 2015 edition (Book 4) [22]. Briefly, after 3 days of adaptive feeding, an injection of 0.5 ml was intraperitoneally administrated every 2 days for three times, and the injection of the same volume was intravenously administrated 2 weeks later to evoke the anaphylactic reaction. The volume dose (0.5 ml) of different injections was according to the CP of 2015 edition, and its solid dose can be calculated by its solid content. The symptoms appeared 30 min after the evocation was observed. The scores were recorded according to Table 1. A higher score suggested the TCMI’s severer toxicity. If an animal showed two symptoms of piloerection, tremble, frequent scratching nose, sneezing three times in a row, three coughs in a row, retching, cyanosis, and dyspnea, or showed any symptom of gatism, unsteady walking or falls, spasm or convulsion, shock, and death, the animal was regarded as the positive animal, and the injection was determined as an unqualified one. Table 1 Scores for symptoms of anaphylactic and anaphylactoid reactions Score . Symptom . Score . Symptom . 1 Piloerection 8 Dyspnea 2 Tremble 9 Gatism 3 Frequent scratching nose 10 Unsteady walking or falls 4 Sneezing three times in a row 11 Spasm or convulsion 5 Three coughs in a row 12 Shock 6 Retching 13 Death 7 Cyanosis Score . Symptom . Score . Symptom . 1 Piloerection 8 Dyspnea 2 Tremble 9 Gatism 3 Frequent scratching nose 10 Unsteady walking or falls 4 Sneezing three times in a row 11 Spasm or convulsion 5 Three coughs in a row 12 Shock 6 Retching 13 Death 7 Cyanosis Open in new tab Table 1 Scores for symptoms of anaphylactic and anaphylactoid reactions Score . Symptom . Score . Symptom . 1 Piloerection 8 Dyspnea 2 Tremble 9 Gatism 3 Frequent scratching nose 10 Unsteady walking or falls 4 Sneezing three times in a row 11 Spasm or convulsion 5 Three coughs in a row 12 Shock 6 Retching 13 Death 7 Cyanosis Score . Symptom . Score . Symptom . 1 Piloerection 8 Dyspnea 2 Tremble 9 Gatism 3 Frequent scratching nose 10 Unsteady walking or falls 4 Sneezing three times in a row 11 Spasm or convulsion 5 Three coughs in a row 12 Shock 6 Retching 13 Death 7 Cyanosis Open in new tab As for detection of anaphylactoid reaction, animals were intravenously administrated with 0.5-ml injection, and the symptoms appeared 30 min after the evocation were observed. The scores were also recorded according to Table 1. Ultraviolet Spectrum of Macromolecules The lyophilized powder of macromolecular substances was diluted with ultrapure water to 100 mg/ml. An aliquot of 100 μl was added into a well of 96-well UV-plate and mixed with pure water of 100 μl. The ultraviolet spectrum of the well was scanned from 230–400 nm with the micro-plate reader. If the maximal absorbance value was bigger than 3 or overflowed, solution of 100 μl in the well was transferred to a new one and double diluted, and its ultraviolet spectrum was scanned again. The maximal absorption wavelength was served as the optimal wavelength in the following size-exclusive chromatography to detect macromolecular signals. Size-Exclusive Chromatography The size-exclusive chromatographic column and a manual injector with a 20-μl sample loop were connected with the pump, respectively. The UV detector was used to detect the absorbance of the effluent from the column. Since all the maximum absorption wavelengths of the macromolecular substances from the six TCMIs were 230 nm, the wavelength was accepted as the detection wavelength throughout the chromatography. The Z-form pool was 5-mm long (3.2 μl). Mobile phase was a mixture of sodium phosphate buffer (50 mM) with 3.5% ammonium sulfate (pH = 7.0) with water (95:5) and eluted at 0.8 ml/min. The temperature of the chromatographic column was kept at 20 ± 1°C. The chromatogram was recorded for 60 min after sample injection (20 μl). The peak area was calculated by the system in an automatic integration mode. In order to observe the recovery and variable coefficient (CV) of the HPLC method, the prepared macromolecular substances of 31.25 μg were added into its original TCMI solution of 1 mL, and the recovery and its CV was calculated after chromatography. The detection limit was obtained by diluting the prepared macromolecular substances until their signal (peak height) at 10 min was two times or more over the noises in the chromatogram. Proteins and Tannins Detected by the Improved Methods Derived from China Pharmacopoeia Proteins and tannins in the original TCMIs and their TCMIs-F were detected by methods documented in CP of 2015 edition and the improved methods. Based on those documented in CP of 2015 edition (Book 4) [23], the improved methods were modified by following an extra step of high-speed centrifugation. Briefly, as for protein detection, sample of 1 ml was mixed with 30% sulfosalicylic acid solution of 1 ml to observe cloud in 5 min (method in CP). Then, the mixture was spun at 10,000 rpm for 10 min. If there was visible precipitate at the bottom of the tube, a positive result was obtained (improved protein detection). As for tannin detection, a sample of 1 ml was mixed with 1% egg white solution (containing 0.9% saline) of 5 ml to observe cloud in 10 min (method in CP). The mixture was also spun at 10,000 rpm for 10 min. If there was visible precipitate at the bottom of the tube, a positive result was also obtained (improved tannin detection). The sample with positive result would be double diluted until a negative result appeared, and the dilution multiple(s) with positive result was/were recorded to measure the relative amount. Statistical Analyses Values were expressed as mean ± SD (standard deviation). One-way analysis of variance (ANOVA) was performed to compare means of measurement data. Statistical significance was accepted at P < 0.05. Results Macromolecular Substances Prepared Macromolecular substances were successfully isolated from original TCMIs (Table 2), suggesting that all the TCMIs containing substances with molecular weight bigger than 10 kDa. Among them, DZXX had the most macromolecular substances, and SHL had the least. Table 2 Substances with molecular weight bigger than 10 kDa were prepared Injection . a Volume (ml) . b Net (mg) . b/a Content (mg/ml) . λmax (nm) . |${E}_{1\mathrm{cm}}^{1\%}$| . DS 20.0 33.80 1.69 230 11.44 DZXX 20.0 107.0 5.35 230 4.63 HH 20.0 12.90 0.65 230 37.86 QKL 20.0 26.10 1.31 230 11.96 SHL 20.0 3.100 0.15 230 6.71 SXN 20.0 51.70 2.59 230 0.23 Injection . a Volume (ml) . b Net (mg) . b/a Content (mg/ml) . λmax (nm) . |${E}_{1\mathrm{cm}}^{1\%}$| . DS 20.0 33.80 1.69 230 11.44 DZXX 20.0 107.0 5.35 230 4.63 HH 20.0 12.90 0.65 230 37.86 QKL 20.0 26.10 1.31 230 11.96 SHL 20.0 3.100 0.15 230 6.71 SXN 20.0 51.70 2.59 230 0.23 Open in new tab Table 2 Substances with molecular weight bigger than 10 kDa were prepared Injection . a Volume (ml) . b Net (mg) . b/a Content (mg/ml) . λmax (nm) . |${E}_{1\mathrm{cm}}^{1\%}$| . DS 20.0 33.80 1.69 230 11.44 DZXX 20.0 107.0 5.35 230 4.63 HH 20.0 12.90 0.65 230 37.86 QKL 20.0 26.10 1.31 230 11.96 SHL 20.0 3.100 0.15 230 6.71 SXN 20.0 51.70 2.59 230 0.23 Injection . a Volume (ml) . b Net (mg) . b/a Content (mg/ml) . λmax (nm) . |${E}_{1\mathrm{cm}}^{1\%}$| . DS 20.0 33.80 1.69 230 11.44 DZXX 20.0 107.0 5.35 230 4.63 HH 20.0 12.90 0.65 230 37.86 QKL 20.0 26.10 1.31 230 11.96 SHL 20.0 3.100 0.15 230 6.71 SXN 20.0 51.70 2.59 230 0.23 Open in new tab Figure 1 Open in new tabDownload slide Ultraviolet spectra of six TCMIs. The spectrum was scanned by a multiple microplate reader from 230 to 400 nm. The volume of the sample in the well was 200 μL, and the concentrations were double diluted to assure their maximum absorption was less than 3. Figure 1 Open in new tabDownload slide Ultraviolet spectra of six TCMIs. The spectrum was scanned by a multiple microplate reader from 230 to 400 nm. The volume of the sample in the well was 200 μL, and the concentrations were double diluted to assure their maximum absorption was less than 3. Table 3 Solid content in TCMIs (mg/ml, n = 3)* Injections . TCMI-R . TCMI-F . Original . DS 22.29 ± 1.09 20.66 ± 1.34 21.83 ± 1.19 DZXX 14.53 ± 0.89 13.96 ± 0.90 13.95 ± 0.88 QKL 40.46 ± 2.10 39.26 ± 1.07 40.34 ± 2.02 SHL 18.72 ± 0.74 18.24 ± 1.83 18.37 ± 0.97 Injections . TCMI-R . TCMI-F . Original . DS 22.29 ± 1.09 20.66 ± 1.34 21.83 ± 1.19 DZXX 14.53 ± 0.89 13.96 ± 0.90 13.95 ± 0.88 QKL 40.46 ± 2.10 39.26 ± 1.07 40.34 ± 2.02 SHL 18.72 ± 0.74 18.24 ± 1.83 18.37 ± 0.97 TCMI-R, injection rich in macromolecules bigger than 10 kDa; TCMI-F, injection free of macromolecules bigger than 10 kDa; and Original, original injection. *No significant difference was found between Original TCMI and its TCMI-R and TCMI-F. Open in new tab Table 3 Solid content in TCMIs (mg/ml, n = 3)* Injections . TCMI-R . TCMI-F . Original . DS 22.29 ± 1.09 20.66 ± 1.34 21.83 ± 1.19 DZXX 14.53 ± 0.89 13.96 ± 0.90 13.95 ± 0.88 QKL 40.46 ± 2.10 39.26 ± 1.07 40.34 ± 2.02 SHL 18.72 ± 0.74 18.24 ± 1.83 18.37 ± 0.97 Injections . TCMI-R . TCMI-F . Original . DS 22.29 ± 1.09 20.66 ± 1.34 21.83 ± 1.19 DZXX 14.53 ± 0.89 13.96 ± 0.90 13.95 ± 0.88 QKL 40.46 ± 2.10 39.26 ± 1.07 40.34 ± 2.02 SHL 18.72 ± 0.74 18.24 ± 1.83 18.37 ± 0.97 TCMI-R, injection rich in macromolecules bigger than 10 kDa; TCMI-F, injection free of macromolecules bigger than 10 kDa; and Original, original injection. *No significant difference was found between Original TCMI and its TCMI-R and TCMI-F. Open in new tab All the macromolecular substances in every TCMI had its ultraviolet spectrum. However, according to Fig. 1, their maximal absorption wavelength (λmax) was 230 nm though there were lower absorption peaks after the wavelength. The wavelength of 230 nm was used in size-exclusive chromatography to detect absorbance signals of macromolecular substances. As expected, the macromolecular substances from different TCMIs had different absorption coefficients (⁠|${E}_{1\mathrm{cm}}^{1\%}$|⁠, Table 2). Anaphylactic and Anaphylactoid Reactions Detected in Guinea Pigs Because the quantity of two TCMIs (SXN and HH) was not enough for animal experiments, only four of them (QKL, SHL, DS and DZXX) were evaluated. Original TCMI TCMI-R and TCMI-F were prepared and their solid content was determined by lyophilization (Table 3). The solid content in TCMI-R was slightly higher, and that in TCMI-F was slightly lower than that in original TCMI; but there was no significance (Student’s t-test) between them. Their solid dose can be calculated from Table 3, also suggesting that there was no significant difference of solid dose between original TCMI, TCMI-R and TCMI-F. Their similar dose made their toxicity comparable. All TCMIs-R got the highest scores, suggesting they were able to cause obvious anaphylactic and anaphylactoid reactions (Table 4) and resulted in the most positive animal(s) (Table 5). The original TCMIs and their TCMIs-F caused almost no anaphylactic and anaphylactoid reactions, if did, much weaker than their TCMIs-R (Table 4). According to the standard of CP, except DZXX, most TCMIs-R were unqualified (Table 5). Table 4 Total score(s) calculated from a group (n = 6) in anaphylactic or anaphylactoid reaction caused by injections Injections . Anaphylactic reaction . Anaphylactoid reaction . TCMI-R . TCMI-F . Original . Negative . TCMI-R . TCMI-F . Original . Negative . DS 16 5 2 0 5 2 1 0 DZXX 21 0 3 0 5 0 1 0 QKL 34 6 8 1 11 2 3 0 SHL 6 0 5 0 3 0 2 0 Injections . Anaphylactic reaction . Anaphylactoid reaction . TCMI-R . TCMI-F . Original . Negative . TCMI-R . TCMI-F . Original . Negative . DS 16 5 2 0 5 2 1 0 DZXX 21 0 3 0 5 0 1 0 QKL 34 6 8 1 11 2 3 0 SHL 6 0 5 0 3 0 2 0 TCMI-R, injection rich in macromolecules bigger than 10 kDa; TCMI-F, injection free of macromolecules bigger than 10 kDa; Original, original injection; and Negative, negative control, namely, normal saline for injection. Open in new tab Table 4 Total score(s) calculated from a group (n = 6) in anaphylactic or anaphylactoid reaction caused by injections Injections . Anaphylactic reaction . Anaphylactoid reaction . TCMI-R . TCMI-F . Original . Negative . TCMI-R . TCMI-F . Original . Negative . DS 16 5 2 0 5 2 1 0 DZXX 21 0 3 0 5 0 1 0 QKL 34 6 8 1 11 2 3 0 SHL 6 0 5 0 3 0 2 0 Injections . Anaphylactic reaction . Anaphylactoid reaction . TCMI-R . TCMI-F . Original . Negative . TCMI-R . TCMI-F . Original . Negative . DS 16 5 2 0 5 2 1 0 DZXX 21 0 3 0 5 0 1 0 QKL 34 6 8 1 11 2 3 0 SHL 6 0 5 0 3 0 2 0 TCMI-R, injection rich in macromolecules bigger than 10 kDa; TCMI-F, injection free of macromolecules bigger than 10 kDa; Original, original injection; and Negative, negative control, namely, normal saline for injection. Open in new tab Table 5 Positive animal(s) appeared in anaphylactic or anaphylactoid reaction caused by injections (n = 6) Injections . Anaphylactic reaction . Anaphylactoid reaction . TCMI-R . TCMI-F . Original . Negative . TCMI-R . TCMI-F . Original . Negative . DS 2 0 0 0 1 0 0 0 DZXX 1 0 0 0 0 0 0 0 QKL 2 0 0 0 1 0 0 0 SHL 2 0 0 0 1 0 0 0 Injections . Anaphylactic reaction . Anaphylactoid reaction . TCMI-R . TCMI-F . Original . Negative . TCMI-R . TCMI-F . Original . Negative . DS 2 0 0 0 1 0 0 0 DZXX 1 0 0 0 0 0 0 0 QKL 2 0 0 0 1 0 0 0 SHL 2 0 0 0 1 0 0 0 TCMI-R, injection rich in macromolecules bigger than 10 kDa; TCMI-F, injection free of macromolecules bigger than 10 kDa; Original, original injection; and Negative, negative control, namely, normal saline for injection. Open in new tab Table 5 Positive animal(s) appeared in anaphylactic or anaphylactoid reaction caused by injections (n = 6) Injections . Anaphylactic reaction . Anaphylactoid reaction . TCMI-R . TCMI-F . Original . Negative . TCMI-R . TCMI-F . Original . Negative . DS 2 0 0 0 1 0 0 0 DZXX 1 0 0 0 0 0 0 0 QKL 2 0 0 0 1 0 0 0 SHL 2 0 0 0 1 0 0 0 Injections . Anaphylactic reaction . Anaphylactoid reaction . TCMI-R . TCMI-F . Original . Negative . TCMI-R . TCMI-F . Original . Negative . DS 2 0 0 0 1 0 0 0 DZXX 1 0 0 0 0 0 0 0 QKL 2 0 0 0 1 0 0 0 SHL 2 0 0 0 1 0 0 0 TCMI-R, injection rich in macromolecules bigger than 10 kDa; TCMI-F, injection free of macromolecules bigger than 10 kDa; Original, original injection; and Negative, negative control, namely, normal saline for injection. Open in new tab Chromatograms of Prepared Macromolecular Substances The prepared macromolecular substances were detected by size-exclusive chromatography, and the chromatograms are shown in Fig. 2. The sample of 20 μl containing 1.250 μg macromolecular substances was injected. All the samples showed that their peaks appeared at about 10 min. The macromolecular substances from HH had the highest absorption coefficient, which caused the highest peak and biggest peak area (Fig. 2). Figure 2 Open in new tabDownload slide Size-exclusive chromatogram of the prepared macromolecular substances (1.250 μg in 20 μl) from six TCMIs. Figure 2 Open in new tabDownload slide Size-exclusive chromatogram of the prepared macromolecular substances (1.250 μg in 20 μl) from six TCMIs. Standard Curve for Quantitative Equation The prepared macromolecular substances were served as the reference to obtain their quantitative equations. By double diluting the prepared macromolecular substances, different peak areas at 10 min were obtained, and the quantitative curves were linearly regressed by the linear equation y = a x (Table 6). Most quantitative equations had good R2 (>0.99), but the equation for DZXX was not as good as others (R2 = 0.9852). Table 3 showed that the method of size-exclusive chromatography has acceptable recoveries, CV and detection limit to quantitatively analyze macromolecular substances in TCMIs. Macromolecular Substances in the TCMIs Macromolecular substances in the original TCMI solution were detected, and their amount was calculated by the quantitative equation shown in Table 6. As expected, all the macromolecular substances in the six TCMIs were positively detected (Fig. 3, Table 7). Among them, the macromolecular substances in HH had the highest peak at 10 min, and then the baseline was kept at a higher level and slowly declined to zero, suggesting that the macromolecular substances in the injection had smaller substances with almost continuous molecular weights less than 10 kDa. As for SXN, there was an obvious peak at about 20 min, suggesting that SXN contained other components with molecular weight less than 10 kDa. If the substances with molecular weight bigger than 10 kDa were removed, the peak at 10 min would dramatically decrease in the chromatograms (Fig. 4), and the content of substances bigger than 10 kDa also dramatically decreased (Table 7). However, the baseline after 10 min was not changed, since the 10 kDa ultra-filter cannot remove substances with molecular weight less than 10 kDa. Proteins and Tannins Detected in the Original TCMIs and TCMIs-F TCMIs are extractions from plants and/or animals and could contain proteins and tannins. Proteins are easily precipitated by sulfosalicylic acid, and tannins will be done so by egg white. As expected, proteins and tannins in the original TCMIs were negatively detected by the methods documented by CP; so were those in their TCMIs-F. However, if proteins and tannins in the original TCMIs and in their TCMIs-F were detected by the improved methods, some positive results appeared. Because the original TCMI contained more macromolecular substances, the proteins and tannins in them would be more likely precipitated by sulfosalicylic acid and egg white, respectively (Fig. 5). As for their TCMIs-F, the proteins and tannins in them were negatively detected or positively detected with lower dilution multiples than those in the original TCMIs. As shown in Fig. 3 and Table 7, HH contained the most macromolecular substances. However, proteins and tannins in both original HH and ‘macromolecule-free’ HH were negatively detected, suggesting that the main macromolecular substances in HH do not belong to protein and tannin. Table 6 Linear equations (y = a x) from the prepared macromolecular substances for TCMI quantitative analysis Injection . Linear regression* (n = 6 ~ 11) . Recovery (%) (n = 3) . CV (%) (n = 3) . Detection limit (ng/20 μl) . . linear range (ng/20 μl) . A . R2 . . . . DS 0.0382–5000 4.185E−05 0.9989 94.7 3.75 0.010 DZXX 0.3052–20,000 4.763E−04 0.9852 117.3 5.25 0.076 HH 0.0012–1250 7.776E−06 0.9972 99.8 3.23 0.001 QKL 0.0048–2500 2.227E−05 0.9937 113.9 2.23 0.001 SHL 0.0191–10,000 6.274E−05 0.9997 92.5 2.68 0.005 SXN 0.3052–10,000 8.847E−04 0.9989 99.6 4.46 0.076 Injection . Linear regression* (n = 6 ~ 11) . Recovery (%) (n = 3) . CV (%) (n = 3) . Detection limit (ng/20 μl) . . linear range (ng/20 μl) . A . R2 . . . . DS 0.0382–5000 4.185E−05 0.9989 94.7 3.75 0.010 DZXX 0.3052–20,000 4.763E−04 0.9852 117.3 5.25 0.076 HH 0.0012–1250 7.776E−06 0.9972 99.8 3.23 0.001 QKL 0.0048–2500 2.227E−05 0.9937 113.9 2.23 0.001 SHL 0.0191–10,000 6.274E−05 0.9997 92.5 2.68 0.005 SXN 0.3052–10,000 8.847E−04 0.9989 99.6 4.46 0.076 *y, macromolecular substances (ng/20 μl); x, peak area (mAu·s). Open in new tab Table 6 Linear equations (y = a x) from the prepared macromolecular substances for TCMI quantitative analysis Injection . Linear regression* (n = 6 ~ 11) . Recovery (%) (n = 3) . CV (%) (n = 3) . Detection limit (ng/20 μl) . . linear range (ng/20 μl) . A . R2 . . . . DS 0.0382–5000 4.185E−05 0.9989 94.7 3.75 0.010 DZXX 0.3052–20,000 4.763E−04 0.9852 117.3 5.25 0.076 HH 0.0012–1250 7.776E−06 0.9972 99.8 3.23 0.001 QKL 0.0048–2500 2.227E−05 0.9937 113.9 2.23 0.001 SHL 0.0191–10,000 6.274E−05 0.9997 92.5 2.68 0.005 SXN 0.3052–10,000 8.847E−04 0.9989 99.6 4.46 0.076 Injection . Linear regression* (n = 6 ~ 11) . Recovery (%) (n = 3) . CV (%) (n = 3) . Detection limit (ng/20 μl) . . linear range (ng/20 μl) . A . R2 . . . . DS 0.0382–5000 4.185E−05 0.9989 94.7 3.75 0.010 DZXX 0.3052–20,000 4.763E−04 0.9852 117.3 5.25 0.076 HH 0.0012–1250 7.776E−06 0.9972 99.8 3.23 0.001 QKL 0.0048–2500 2.227E−05 0.9937 113.9 2.23 0.001 SHL 0.0191–10,000 6.274E−05 0.9997 92.5 2.68 0.005 SXN 0.3052–10,000 8.847E−04 0.9989 99.6 4.46 0.076 *y, macromolecular substances (ng/20 μl); x, peak area (mAu·s). Open in new tab Figure 3 Open in new tabDownload slide Size-exclusive chromatograms of six original TCMIs. Figure 3 Open in new tabDownload slide Size-exclusive chromatograms of six original TCMIs. Table 7 Macromolecular substances in the original TCMIs and their TCMIs-F* Injection . Original . TCMI-F . Peak area (mAu·s) . Macromolecules (ng/20 μl) . Macromolecules (ng/ml) . Peak area (mAu·s) . Macromolecules (ng/20 μl) . Macromolecules (ng/ml) . DS 11,327.3 0.474 23.7 1533.4 0.064 3.2 DZXX 1310.8 0.624 31.2 291.2 0.139 6.9 HH 152,654.9 1.187 59.4 31,348.1 0.244 12.2 QKL 10,156.3 0.226 11.3 4710.5 0.105 5.2 SHL 3359.5 0.211 10.5 495.6 0.031 1.6 SXN 850.5 0.752 37.6 275.7 0.244 12.2 Injection . Original . TCMI-F . Peak area (mAu·s) . Macromolecules (ng/20 μl) . Macromolecules (ng/ml) . Peak area (mAu·s) . Macromolecules (ng/20 μl) . Macromolecules (ng/ml) . DS 11,327.3 0.474 23.7 1533.4 0.064 3.2 DZXX 1310.8 0.624 31.2 291.2 0.139 6.9 HH 152,654.9 1.187 59.4 31,348.1 0.244 12.2 QKL 10,156.3 0.226 11.3 4710.5 0.105 5.2 SHL 3359.5 0.211 10.5 495.6 0.031 1.6 SXN 850.5 0.752 37.6 275.7 0.244 12.2 *TCMI-F was made by removing substances whose molecular weight was bigger than 10 kDa. Calculated from Figs 3 and 4. Open in new tab Table 7 Macromolecular substances in the original TCMIs and their TCMIs-F* Injection . Original . TCMI-F . Peak area (mAu·s) . Macromolecules (ng/20 μl) . Macromolecules (ng/ml) . Peak area (mAu·s) . Macromolecules (ng/20 μl) . Macromolecules (ng/ml) . DS 11,327.3 0.474 23.7 1533.4 0.064 3.2 DZXX 1310.8 0.624 31.2 291.2 0.139 6.9 HH 152,654.9 1.187 59.4 31,348.1 0.244 12.2 QKL 10,156.3 0.226 11.3 4710.5 0.105 5.2 SHL 3359.5 0.211 10.5 495.6 0.031 1.6 SXN 850.5 0.752 37.6 275.7 0.244 12.2 Injection . Original . TCMI-F . Peak area (mAu·s) . Macromolecules (ng/20 μl) . Macromolecules (ng/ml) . Peak area (mAu·s) . Macromolecules (ng/20 μl) . Macromolecules (ng/ml) . DS 11,327.3 0.474 23.7 1533.4 0.064 3.2 DZXX 1310.8 0.624 31.2 291.2 0.139 6.9 HH 152,654.9 1.187 59.4 31,348.1 0.244 12.2 QKL 10,156.3 0.226 11.3 4710.5 0.105 5.2 SHL 3359.5 0.211 10.5 495.6 0.031 1.6 SXN 850.5 0.752 37.6 275.7 0.244 12.2 *TCMI-F was made by removing substances whose molecular weight was bigger than 10 kDa. Calculated from Figs 3 and 4. Open in new tab Discussion and Conclusion The safety of four injections (QKL, SHL, DS and DZXX) in anaphylactic and anaphylactoid reactions was verified in guinea pigs. TCMIs-R were more likely to cause anaphylactic and anaphylactoid reactions (Tables 4 and 5). Without sensitization, TCMIs-R intravenously injected can also cause symptoms. Therefore, the symptoms should belong to anaphylactoid reactions rather than anaphylactic reaction. The results supported that, macromolecular substances were a very dangerous factor associated with anaphylactic and anaphylactoid reactions, suggesting that macromolecular substances in TCMIs should be removed as thoroughly as possible. And it is urgent to establish a sensitive method to detect them. Size-exclusive chromatography is a regular method to determine macromolecular substances in injections [24], but it is a big challenge to use the method to analyze macromolecules in TCMIs since there were rare successful reports other than DS [21]. The present study tried to apply the method to analyze the macromolecular substance. In the present study, the macromolecular substances were successfully isolated from TCMIs by ultra-filtration, and the substances in the original TCMIs can be separated and detected by size-exclusive chromatography. The present study also proved that, the isolated macromolecular substances were dark brown, and the color of its TCMI-F was much lighter than that of the original one [25]. However, the amount of macromolecular substances isolated from a TCMI was much more than those detected in the original one, suggesting that some macromolecular substances could be generated during preparation. TCMIs are biological extracts, possibly containing phenols and tannins. Both phenols and tannins are sensitive to oxygen and can be easily polymerized and transformed to macromolecular substances [26, 27]. Thus, part of the isolated macromolecular substances could not be those in the original TCMI. The results also suggested that isolation of original macromolecular substances is a tough task. Because there were no better reference substances, we had to use them as a reference to evaluate the macromolecular substances in TCMIs. Figure 4 Open in new tabDownload slide Size-exclusive chromatograms of six TCMIs ‘free’ of macromolecules (TCMIs-F). TCMI-F was made by removing substances with molecular weight bigger than 10 kDa. Figure 4 Open in new tabDownload slide Size-exclusive chromatograms of six TCMIs ‘free’ of macromolecules (TCMIs-F). TCMI-F was made by removing substances with molecular weight bigger than 10 kDa. Figure 5 Open in new tabDownload slide Proteins (A) and tannins (B) were detected by improved methods similar to China Pharmacopeia (2015 edition) but followed with an extra centrifugation. Figure 5 Open in new tabDownload slide Proteins (A) and tannins (B) were detected by improved methods similar to China Pharmacopeia (2015 edition) but followed with an extra centrifugation. By coincidence, macromolecular substances isolated from TCMIs used in the present study had ultraviolet absorption. Because the micro-plate reader cannot scan the absorbance at wavelengths shorter than 230 nm, the wavelength of 230 nm was used to detect signals in the chromatography. According to the ultraviolet spectra, it can be deduced that the absorbance at 190–230 nm should be much stronger than that at 230 nm. Therefore, a shorter wavelength would be more sensitive in detecting absorbance signals. Nevertheless, absorbance at 230 nm was strong enough to detect macromolecular signals, and the macromolecular substances from different TCMIs had different absorption coefficients, suggesting that the macromolecular substances are different from each other. The macromolecular substances in the TCMIs could be a group of macromolecules with different molecular weights. The Sephadex G-10 column used in the present study is able to separate macromolecular substances from 4–150 kDa [28]. The substances of 10 kDa or more in TCMIs were eluted at about 10 min. The substances of less than 10 kDa were eluted later. As for the original TCMIs, since the relative small molecules could have almost continuous molecular weights, there were no obvious peaks after 10 min, but the base lines were elevated (Fig. 3) except SXN. SXN was made of flavonoids from ginkgo leaves [29], the flavonoids must make the peak at about 20 min (Figs 3 and 4). As for TCMIs-F, the peaks at 10 min were reduced, which indirectly proved that the macromolecular substances can be removed by ultra-filtration (Fig. 4). Because of the zoomed-in ordinate of the chromatogram, the small peaks after 10 min became manifested (Fig. 4). It should be pointed out that the quantitative equation for macromolecular substances in TCMIs was not as good as those for impurities in other injections like cephalosporin for injection [24]. The reason could be associated with the chemical complexity of the TCMIs concerned. The six TCMIs were made of the total extract from one or more natural raw drug(s) and could include more than one hundred compounds known and numerous compounds unknown. The chemical complexity would affect the efficiency of size-exclusive chromatography and resulted in a relative ‘bad’ quantitative equation. TCMIs used in this present study were extracted from herbal raw drugs, and there could be proteins and tannins in them. The proteins and tannins in TCMIs can be tested by sulfosalicylic acid solution and egg white solution [18, 19]. The proteins and tannins in TCMIs-R would be more likely to be positively tested. Instead, those in TCMIs-F were more likely to be negatively tested (Fig. 5). The results were consistent with those of size-exclusive chromatography (Figs 3 and 4, and Table 7). Obviously, the results of size-exclusive chromatography were much objective and quantifiable to evaluate macromolecular substances. Taken together, the present study verified that macromolecular substances in TCMIs are a dangerous factor causing safety problems including anaphylactic and anaphylactoid reactions and provided a chromatographic method to analyze them in TCMIs. Funding The authors acknowledge the financial support from the National Natural Science Foundation of China (NSFC 81560645) and the Yunnan Provincial Science and Technology Department–Applied Basic Research Joint Special Funds of Yunnan University of Chinese Medicine (2017FF117-032). Conflict of interest The authors report no conflicts of interest. The authors alone are fully responsible for the content and writing of this article. References 1. Kong X-Y , Hao Y, Wu T-X, Xie YM. Adverse drug reactions or adverse events of Chaihu injection: a systematic review . Chin J Integr Med 2010 ; 8 : 1124 – 32 . Google Scholar Crossref Search ADS WorldCat 2. Yin H , Ke J, Duan W. Macromolecules in traditional Chinese medicine injections: active components or impurities? Negative 2018 ; 9 : 50 – 5 . OpenURL Placeholder Text WorldCat 3. He P , Li F, Li L et al. Developing traditional Chinese medicine injection is the need for curing sickness to save patients . China Journal of Chinese Materia Medica 2017 ; 42 : 1011 – 4 . Google Scholar PubMed OpenURL Placeholder Text WorldCat 4. Yuan T , Chen Y, Zhou X et al. Effectiveness and safety of Danshen injection on heart failure: protocol for a systematic review and meta-analysis . Medicine (Baltimore) 2019 ; 98 : e15636 . Google Scholar Crossref Search ADS PubMed WorldCat 5. Li Y , Lin G, Xie Y et al. Postmarketing evaluation on the safety and effectiveness of deng-zhanxixin injection made from Dengzhanxixin (Herba Erigerontis Breviscapi) . J Tradit Chin Med 2015 ; 35 : 99 – 103 . Google Scholar Crossref Search ADS PubMed WorldCat 6. Feng WW , Zhang Y, Tang JF et al. Combination of chemical fingerprinting with bioassay, a preferable approach for quality control of safflower injection . Anal Chim Acta 2018 ; 1003 : 56 – 63 . Google Scholar Crossref Search ADS PubMed WorldCat 7. Li Q , Zhang L, Gu L et al. Pseudo-allergic reaction caused by Qingkailing injection partially via the PI3K-Rac1 signaling pathway in RBL-2H3 cells . Toxicol Res (Camb) 2019 ; 8 : 353 – 60 . Google Scholar Crossref Search ADS PubMed WorldCat 8. Gao Y , Hou R, Han Y et al. Shuang-Huang-Lian injection induces an immediate hypersensitivity reaction via C5a but not IgE . Sci Rep 2018 ; 8 : 3572 . Google Scholar Crossref Search ADS PubMed WorldCat 9. Wang C , Shi QP, Ding F et al. Reevaluation of the post-marketing safety of Shuxuening injection based on real-world and evidence-based evaluations . Drug Des Devel Ther 2018 ; 12 : 757 – 67 . Google Scholar Crossref Search ADS PubMed WorldCat 10. Wang Y , Fan LP, Song J et al. Retrospective analysis and discussion on 74 cases of adverse reactions of traditional Chinese medicine injection . China Journal of Chinese Materia Medica 2018 ; 43 : 4347 – 51 . Google Scholar PubMed OpenURL Placeholder Text WorldCat 11. Tan L-J , Wang M, Zhu Y. Research progress of adverse reactions of traditional Chinese medicine injections . China J Chinese Materia Medica 2014 ; 39 : 3889 – 998 . OpenURL Placeholder Text WorldCat 12. Xiang F , Zhang X. Chinese medicine adverse reactions' literature statistical analysis in recent five years . China Journal of Chinese Materia Medica 2011 ; 36 : 2755 – 8 . Google Scholar PubMed OpenURL Placeholder Text WorldCat 13. Wang ZF , Ai QH, Li YY et al. Analysis of the allergic reaction types of Chinese medicine injection based on immunotoxicty . China Journal of Chinese Materia Medica 2015 ; 40 : 4762 – 5 . Google Scholar PubMed OpenURL Placeholder Text WorldCat 14. Wang H , Zhang X, Wang D et al. An on-line analytical approach for detecting haptens in Shuang-huang-lian powder injection . J Chromatogr A 2017 ; 1513 : 126 – 39 . Google Scholar Crossref Search ADS PubMed WorldCat 15. Yin H , Ke J, Duan W. Are macromolecules in traditional Chinese medicine injections active components or impurities? Negative 2018 ; 9 : 26 – 33 . OpenURL Placeholder Text WorldCat 16. Ke J , Zhang L, Yin H et al. Macromolecular substances from traditional Chinese medicine injection caused safety problems . Chinese Traditional Patent Medicine 2014 ; 36 : 855 – 9 . OpenURL Placeholder Text WorldCat 17. Duan W , Zhang L. The strategies to increase the safety of traditional Chinese medicine injections . Chinese Traditional Patent Medicine 2012 ; 34 : 2201 – 5 . OpenURL Placeholder Text WorldCat 18. Duan W , Ke J, Li Q et al. Condensed tannins were detected by PVDF membrane coated with proteins . Chinese Traditional Patent Medicine 2011 ; 33 : 80 – 3 . OpenURL Placeholder Text WorldCat 19. W. Duan , Q. Ling, K. Jin et al. Detection of trace protein in Chinese Materia Medica injections by soaking PVDF membrane. In: 2011 International Conference on Remote Sensing, Environment and Transportation Engineering Proceedings , 2011 , 6718 – 20 . 20. Liu Q , Yin H, Hou X et al. The improved method to detect resins in traditional Chinese medicine . Chinese Traditional Patent Medicine 2016 ; 38 : 1872 – 4 . OpenURL Placeholder Text WorldCat 21. Lai Y , Tong H, Zhang X et al. Subtractive analysis of size exclusion chromatography and fingerprint for radix salviae miltiorrhizae injection . Traditional Chinese Drug Research & Clinical Pharmacology 2009 ; 20 : 376 – 9 . OpenURL Placeholder Text WorldCat 22. Chinese Pharmacopoeia Committee . Chinese Pharmacopoeia , 2015 edn, Book 4 . Beijing : China Medical Science Press , 2015 , 159 . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC 23. Chinese Pharmacopoeia Committee . Chinese Pharmacopoeia , 2015 edn, Book 4 . Beijing : China Medical Science Press , 2015 , 225 . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC 24. Xu Y , Wang D, Zhu B et al. Separation and characterization of allergenic polymerized impurities from cephalosporin for injection by trap free two-dimensional high performance size exclusion chromatographyxreversed phase liquid chromatography coupled with ion trap time-of-flight mass spectrometry . J Pharm Biomed Anal 2018 ; 154 : 425 – 32 . Google Scholar Crossref Search ADS PubMed WorldCat 25. Li Y , Duan J, Bai W et al. Influence of macromolecule on appearance quality of traditional Chinese medicine injections . J Tradit Chin Med 2018 ; 24 : 43 – 8 . Google Scholar Crossref Search ADS WorldCat 26. Wang A , Lin L, Wang Y. Traditional Chinese herbal medicine penthorum chinense pursh: a phytochemical and pharmacological review . Am J Chin Med 2015 ; 43 : 601 – 20 . Google Scholar Crossref Search ADS PubMed WorldCat 27. Liu QM , Jiang JG. Antioxidative activities of medicinal plants from TCM . Mini Rev Med Chem 2012 ; 12 : 1154 – 72 . Google Scholar Crossref Search ADS PubMed WorldCat 28. Fuchigami T , Ono H, Oyadomari K et al. Development of a (68)Ge/(68)Ga generator system using polysaccharide polymers and its application in PET imaging of tropical infectious diseases . ACS Omega 2017 ; 2 : 1400 – 7 . Google Scholar Crossref Search ADS PubMed WorldCat 29. Yu S , Li J, Guo L et al. Integrated liquid chromatography-mass spectrometry and nuclear magnetic resonance spectra for the comprehensive characterization of various components in the Shuxuening injection . J Chromatogr A 2019 ; 1599 : 125 – 35 . Google Scholar Crossref Search ADS PubMed WorldCat © The Author(s) 2020. Published by Oxford University Press. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)
TI - Macromolecular substances as a dangerous factor in traditional Chinese medicine injections were determined by size-exclusive chromatography
JF - Toxicology Research
DO - 10.1093/toxres/tfaa024
DA - 2020-06-30
UR - https://www.deepdyve.com/lp/oxford-university-press/macromolecular-substances-as-a-dangerous-factor-in-traditional-chinese-5YMJ2jd0aC
SP - 323
EP - 330
VL - 9
IS - 3
DP - DeepDyve
ER -