Attenuated Structural Transformation of Indaconitine during Sand Frying Process and Anti-Arrhythmic Effects of Its Transformed Products

The transformation pathways of diterpenoid alkaloids have been clarified clearly in the boiling and steaming process, but remain to be determined in the sand frying process. The aims of the study were to investigate the transformation pathways of indaconitine in the sand frying process, as well as examine the cardiotoxicity and anti-arrhythmic activity of indaconitine and its transformed products. The transformed product was separated by column chromatography, and the structure was identified by 1H NMR, 13C NMR, and HR-ESI-MS. The cardiotoxicity of indaconitine and its transformed products was clarified by observing the electrocardiogram (ECG) changes at the same dose. Furthermore, the anti-arrhythmic activity of the transformed products was investigated using an aconitine-induced rat arrhythmia model. Consequently, Δ15(16)-16-demethoxyindaconitine, a new diterpenoid alkaloid, was isolated from processed indaconitine. Intravenous injection of 0.06 mg/kg indaconitine induced arrhythmias in SD rats, while Δ15(16)-16-demethoxyindaconitine did not exhibit arrhythmias at the same dose. In the anti-arrhythmic assay, mithaconitine, obtained in the previous research, together with Δ15(16)-16-demethoxyindaconitine, could dose-dependently delay the onset time of ventricular premature beat (VPB) and reduce the incidence of ventricular tachycardia (VT), combined with the increasing arrhythmia inhibition rate, exhibiting strong anti-arrhythmic activities. These results indicated that two or more pathways exist in the sand frying process, and the transformed products exhibited lower cardiotoxicity and strong anti-arrhythmic activities, which had the possibility of being developed into anti-arrhythmic drugs.

As is well known, the unprocessed Aconitum herbs mainly contain toxic diterpenoid alkaloids [3- 5,8], to ensure safety and effectiveness, and they must be processed before clinical medication. Generally, Aconitum plants are processed by "boiling with water for 4-6 hours" or "steaming for 6-8 hours" in traditional Chinese medicine [5,9], which has been documented in the Chinese Pharmacopoeia 2020. While the sand frying method is usually applied in Qiang and Tibetan medicine, it only takes 10-15 minutes of stir-frying with sand to achieve the purpose of detoxification. During the boiling and steaming process, the structural transformation pathways of diterpenoid alkaloids are relatively conservative, mainly adhering to the following processing principles: the acetoxyl group at C-8 and benzoyl group at C-14 of toxic diterpenoid alkaloids are unstable, easily hydrolyzed in the presence of water, and then converted to the low toxic monoester-and amine-diterpenoid alkaloids [9,10]. For example, the acetoxyl group at C-8 of aconitine is hydrolyzed to obtain benzoylaconine, and the benzoyl group at C-14 of benzoylaconine is further hydrolyzed to generate aconine [9,10]. However, the structural transformation pathways of these components are more complicated during the process of sand frying, and the transformation mechanisms have not been fully elucidated [11].
Mithaconitine, a sand-fried product of indaconitine, has been isolated from processed indaconitine in our previous study [12]. It was also confirmed that the cardiotoxicity was lower than that of the prototype compound indaconitine, which was the main ingredient of A. tatsienense Finet & Gagnep. [12,13]. In this study, another main transformed compound was further separated and identified. Besides, the cardiotoxicity was also investigated. By comparing the changes in structure and toxicity between the prototype compound and the converted products, the structural transformation pathways of indaconitine would be clarified, as well as whether the sand frying method could attenuate the toxicity.
Notably, some diterpenoid alkaloids with similar structures exert opposite effects. For example, aconitine, a famous diterpenoid alkaloid, has fatal arrhythmogenic effects, while its structurally similar alkaloids 14-benzoyltalatisamine, 14-benzoyldelcosine, and 14-benzoylidocarpine [14][15][16] have anti-arrhythmic effects. Since the structure of indaconitine is similar to its transformed products, it is unclear, with the transformation of structure, whether the transformed components will generate an anti-arrhythmic effect on the premise of reduced toxicity. erefore, the antiarrhythmic activity of transformed components was further investigated by employing a rat model of aconitine-induced arrhythmia.

General Experimental
Procedures. BL-420F organism functional experimental system (Chengdu Techman Co., Ltd., Chengdu, China) was used to record the lead II electrocardiograms (ECGs) of rats. CPA2250 electronic analytical balance (Sartorius, Germany) was used to correctly weigh the laboratory samples. HH-SJ heat-collecting magnetic stirrer (Jintan Chengdong Xinrui Instrument Factory, Changzhou, China) was used for simulating the process of sand frying by oil bath heating. Optical rotations were obtained on a PerkinElmer 341 polarimeter (Perki-nElmer, USA). e melting point was measured on an X-4 micro-melting point apparatus and uncorrected (Shanghai Precision Scientific Instrument Corporation, Shanghai, China). NMR spectra were measured on a Bruker Avance 600 spectrometer (Bruker, Germany) in CDCl 3 with tetramethylsilane (TMS) as an internal standard. Mass spectra were carried out on a micrO-TOF-Q-II mass spectrometer (Bruker, Germany

Preparation and Separation of Transformed Component.
520 mg indaconitine was dissolved in a 250 mL roundbottomed flask with a proper amount of dichloromethane, and the solvent was removed in vacuo to make it uniformly adhere to the inner wall of the flask. e flask was subsequently immersed in an oil bath, heated at 160°C for 30 min, and cooled to room temperature after reaction, and a crude residue of indaconitine (450 mg) was obtained for column chromatography.

Electrocardiography.
e rats were anesthetized by intraperitoneal (i.p.) injection of 20% urethane (1.2 g/kg), with their back fixed, and four limbs in subcutaneous penetration of needle electrodes. Lead II ECGs were recorded after the administration of urethane. ECG acquisition was performed using the BL-420F organism functional experimental system.

2
Evidence-Based Complementary and Alternative Medicine 2.6. Cardiotoxicity Test. 20 male and female SD rats were randomly divided into two groups (n � 10 for each group): indaconitine and Δ 15(16) -16-demethoxyindaconitine (compound 1). At the beginning of the experiments, both groups of rats were anesthetized using 20% urethane (1.2 g/kg, i.p.), recording the lead II ECGs for 20 min prior to drug administration, and then, the rats were intravenously injected the same dose of experimental compounds. ECGs were continually conducted for 30 min after drug administration [12] (Figure 1). rough the preliminary experiment, it was found that 0.06 mg/kg indaconitine caused ventricular premature beat (VPB), ventricular tachycardia (VT), and ventricular fibrillation (VF) in normal rats. Whether the converted products could induce arrhythmias at the same dose can directly reflect the structural changes in the prototype compound on its cardiotoxicity.

Aconitine-Induced Arrhythmia Test.
To further investigate the anti-arrhythmic activity of the converted alkaloids, the testing was performed in vivo in 197 rats. e rats were randomly divided into the following 14 groups, and the grouping is shown in Table 1. e blank solvent was configured by taking 4 mL of 1% HCl ethanol and adding saline for volume fixation to 100 mL, the preparation methods of aconitine, positive drugs, and the experimental compounds were the same as the blank solvent. To confirm whether the blank solvent would influence the ECGs of rats, the rats of the blank solvent group were only administered the same volume of blank solvent to observe ECG changes during the recording time. e rats were anesthetized using 20% urethane (1.2 g/kg, i.p.) [19,20]. Recording the lead II ECGs for 20 min prior to administration, then the experimental compounds, positive drugs, and equal volume of saline were subsequently administered via the exposed vena femoral, respectively. After stabilization for 10 min [21], aconitine was administered into the vena femoral at a dosage of 0.03 mg/kg [22][23][24] to establish arrhythmia ( Figure 1). e onset time of VPB [25] was recorded within 30 min [25][26][27] after aconitine injection. Meanwhile, the VT [28,29] or arrhythmia [30], if any, was recorded for each group at the end of observation period.

Statistical Analysis.
e data obtained were analyzed using Statistical Package for Social Sciences (SPSS) version 20 software. Experimental data were expressed as mean ± SD or proportion. Descriptive statistics were examined individually. When the data conformed to the normal distribution, the one-way ANOVA would be used for those with homogeneous variance, if the variance was not homogeneous, Tamhane's T2 test would be used for comparison between groups. Comparisons of proportions were made with Pearson's chi-square (χ 2 ) test. P < 0.05 was considered to be a statistically significant difference.

Structural Identification of the Converted Product.
Compound 1 was isolated from processed indaconitine by procedures as described in the experimental section.   [32,33], resulting in the location of the benzoyl group to C-14. ree methoxyl groups were attributed to C-1, C-6, and C-18 based on the cross-peaks between 1-OCH 3 (δ H 3.23, s) and C-1 (δ C 82.4, d), 6-OCH 3 (δ H 3.14, s) and C-6 (δ C 82.3, d), and 18-OCH 3 (δ H 3.30, s) and C-18 (δ C 76.9, t) in its HMBC spectrum ( Figure 2). Two hydroxyl groups were assigned to C-3 and C-13 based on the correlations between the C-3 (δ C 71.7, d) and H-18 (δ H 3.68, 3.75), H-19 (δ H 2.35, 2.89), as well as C-13 (δ C 76.1, s) and H-9 (δ H 2.95), H-15 (δ H 6.54), in the HMBC of 1. e 1 H NMR signal at δ 6.54 (1H, d, J � 10.26 Hz) and δ 6.06 (1H, d, J � 10.26 Hz) and the 13 C NMR signals at δ 125.5 and 137.1 indicated the existence of a disubstituted double bond (-CH�CH-) in the molecule. In accordance with the spectral data and the general rules for C 19 -diterpenoid alkaloids, the disubstituted double bond could only be at Δ 2(3) or Δ 15 (16) [34][35][36]. As mentioned above, a hydroxyl group has already located in the C-3 position. erefore, the double bond must be at Δ 15 (16) . Also, long-range correlations between C-13 and H-15, and C-8 and H-16 confirmed the presence of a double bond between C-15 and C-16. In addition, the coupling constant between H-15 and H-16 was 10.26 Hz, suggesting that the two hydrogen atoms were CIS structures. e key NOE correlations ( Figure 2) between H-15 and H-17, and H-16 and H-17 also confirmed the configuration of H-15 and H-16 in 1 was α-orientation. Furthermore, the configuration of 8-OAc in 1 was determined to have a β-orientation according to the key NOE correlations between 8-OAc and H-2′, 6′ (Figure 2). e NMR spectra of compound 1 lacked a methoxyl group at C-16, when compared with indaconitine [34], and the 13 C NMR spectra of 1 clearly showed changes in the chemical shifts of C-15 and C-16 due to the presence of a double bond. Except for these points, the 13 C NMR spectra of the two alkaloids were very similar. us, the structure of 1 was elucidated as Δ 15(16) -16-demethoxyindaconitine, a new diterpenoid alkaloid (Figures S1-S14).

Cardiotoxicity of Indaconitine and Its Transformed
Product. Intravenous injection of 0.06 mg/kg indaconitine resulted in disturbances of cardiac rhythm progressing from VPB to VT in all rats, some of the rats developed to VF and     Table 3 and Figure 3.

Effects of the Converted Products on VPB Incubation
Period. VPB is the initial manifestation of aconitine-induced arrhythmia model rats, and VPBs are characterized by premature and bizarrely shaped QRS complexes that appear wide on the ECG, these complexes are not preceded by a P wave, and the T wave is usually oriented in a direction opposite the major deflection of the QRS [37,38]. VPB incubation period refers to the time after the administration of aconitine to the first occurrence of VPB [25]. e length of the VPB incubation period can reflect the anti-arrhythmic effect of the experimental compounds, and the longer the incubation period is, the better the anti-arrhythmic effect is. e typical characteristic ECGs of VPB and VT occurred successively following the administration of aconitine, even VF occurred in parts of the rats, and arrhythmias lasted for more than 30 min, suggesting that the arrhythmia model was successfully established. e VPB incubation period of the control group was 97.7 ± 23.7 s. Lidocaine and propafenone as the positive anti-arrhythmic drugs delayed the onset time of VPB induced by aconitine in rats, and the latency of VPB was 180.9 ± 51.0 s and 142.3 ± 17.5 s, respectively. Two positive drugs had a significant difference in comparison with the control group (P < 0.05). Besides, lidocaine could protect 44.4% of rats from arrhythmia, exhibiting a better anti-arrhythmic effect than propafenone. e effects of mithaconitine and Δ 15(16) -16-demethoxyindaconitine on VPB incubation period in rats are shown in Table 4 and Figure 4. Compared with the control group, the emergence of VPB in different dose groups of mithaconitine was significantly delayed (P < 0.05, Figure 4(a)), except for the 0.05 mg/kg dose group. e VPB incubation periods in 0.15 mg/kg, 0.30 mg/kg, and 0.40 mg/kg groups were 256.6 ± 45.1 s, 527.6 ± 208.6 s, and 583.4 ± 219.7, respectively. ey were significantly different from propafenone (P < 0.05). Similarly, compared with lidocaine, there was also a significant difference in groups of 0.30 mg/kg and 0.40 mg/ kg (P < 0.05), indicating that a 0.30 mg/kg or more intravenous dose of mithaconitine exhibited a marked activity relative to the positive drugs.     dose range of 0.20-0.40 mg/kg exhibited superior anti-arrhythmic activity relative to mithaconitine and the positive drugs.

Effects of the Converted Products on the Incidence of VT.
VT is the result of further development of VPB, a malignant ventricular arrhythmia, as well as an important factor in sudden cardiac death, which can be avoided if VT is corrected promptly and effectively [39]. e incidence of VT can be used to evaluate whether the experimental compounds can effectively prevent the progress of VPB, and the lower the occurrence is, the better the efficacy is.
On the basis of the aforementioned results, it can be concluded that mithaconitine and Δ 15(16) -16-demethoxyindaconitine reduced the incidence of VT in a dose-dependent manner, suggesting that they could effectively prevent the further development of VPB.

Effects of the Converted Products on Arrhythmia Inhibition Rate.
e incidence of arrhythmia [30] is defined as the proportion of arrhythmia that occurs within 30 min after pre-intravenous injections of the test compounds, and an arrhythmia model was established with aconitine. e occurrence of any kind of ECGs should be judged as arrhythmias, such as VPB, VT, or VF. e arrhythmia inhibition rate (%) � 100%-arrhythmia incidence rate (%), namely the proportion that no arrhythmia occurs within 30 min, which can be used to evaluate whether the experimental compounds could completely inhibit the arrhythmogenic effect of aconitine. e arrhythmia inhibition rate is the most intuitive index reflecting the strength of the drug efficacy, and the higher it gets, the better the efficacy is.
Considering that the number of rats without arrhythmia in 0.05 mg/kg, 0.10 mg/kg, and 0.15 mg/kg dose groups of mithaconitine was small, the chi-square test could not be performed according to the original grouping. erefore, the chi-square analysis was carried out by combining the above three groups into 0.05-0.15 mg/kg group. Similarly, the 0.05 mg/kg, 0.10 mg/kg, and 0.20 mg/kg dose groups of Δ 15(16) -16-demethoxyindaconitine were merged to conduct data analysis.

Discussion
Currently, Aconitum medicinal materials are mainly processed by boiling and steaming in traditional Chinese medicine, and the processing mechanisms underlying the above methods are attributed to diester alkaloids being sequentially converted into less toxic monoester-and aminealkaloids through ester bond hydrolysis reaction [9,10]. In our previous study, a similar structural transformation mechanism was found. In the boiling and steaming process,   aconitine, deoxyaconitine, and 3-acetylaconitine, the main ingredients of A. pendulum Busch, were hydrolyzed to generate benzoylaconine, benzoyldeoxyaconine, and polyschistine-D via hydrolysis of the acetoxyl group at C-8 [11,40]. erefore, it could be speculated that indaconitine was sequentially converted into 14-benzoylpseudaconine [41] and pseudoaconine [42] based on the similar transformation mechanism (Figure 7). In our previous study, mithaconitine, a sand-fried transformed product of indaconitine, has been isolated and identified. It also demonstrated that the cardiotoxicity of mithaconitine was attenuated after processing [12]. In this study, another main transformed product Δ 15(16) -16demethoxyindaconitine was further isolated by column chromatography, and the structure was elucidated by NMR and HR-ESI-MS.
By comparing the structural differences between indaconitine and its transformed products, mithaconitine and Δ 15(16) -16-demethoxyindaconitine, it could be seen that the transformation pathways of indaconitine in the sand frying process were different from that in boiling and steaming process. ere might be at least two transformation pathways in the sand frying process: (1) the acetoxyl group at C-8 of indaconitine was firstly hydrolyzed to a hydroxyl group to get 14-benzoylpseudaconine, and then, a double bond at C-8/C-15 was introduced via further dehydration to convert into mithaconitine. (2) Indaconitine was converted into Δ 15(16) -16-demethoxyindaconitine by elimination of the hydrogen atom at C-15 and the methoxyl group at C-16 ( Figure 7). us, the differences between the boiling, steaming, and sand frying methods lay in that, in the first two methods, the acetoxyl group at C-8 of indaconitine was firstly hydrolyzed to a hydroxyl group, and the benzoyl group at C-14 was subsequently hydrolyzed. However, the sand frying method was different. In pathway A, the acetoxyl group at C-8 of indaconitine was firstly hydrolyzed to a hydroxyl group and then formed a double bond at C-8/C-15 through dehydration reaction, but the benzoyl group of C-14 was not changed. In pathway B, the ester bonds at C-8 and C-14 were not changed, but generated a double bond at C-15/C- 16. at is, only hydrolysis reaction occurred in the boiling and steaming process. However, not only hydrolysis reactions but also dehydration reactions occurred in the sand frying process. It is speculated that the reasons for the structural transformation pathways differing among the three processing methods are as follows: (1) the processing temperature is different. e boiling point of water is 100°C under normal atmospheric pressure, and the temperature of water vapor is slightly above 100°C, whereas the gravel can reach 200°C or even higher. (2) e heating medium is different. e medium in contact with the herbs is boiling water or water vapor in the boiling or steaming process, which belongs to the "water and fire processing" [9]. In contrast, the gravel is a solid medium, and there is almost no water involved in the process of sand frying, except for the small amount of moisture contained in the herbs in accordance with the pharmacopoeia, which is more compatible with "fire processing" [43].
e cardiotoxicity test showed that intravenous injection of 0.06 mg/kg indaconitine caused arrhythmias in normal rats, such as VPB, VT, and VF, even 20% of rats died of severe arrhythmias. However, none of the rats in Δ 15(16) -16demethoxyindaconitine and mithaconitine [12] groups showed the above arrhythmias. e cardiotoxicity of the transformed products was reduced compared with the prototype compound indaconitine, indicating that the sand frying method contributed to attenuating the cardiotoxicity.
In this study, three indicators, the VPB incubation period, the incidence of VT, and arrhythmia inhibition rate, were used to comprehensively assess the anti-arrhythmic activity of the transformed products of indaconitine.
e experimental results manifested that, in the respective dose ranges, mithaconitine and Δ 15(16) -16demethoxyindaconitine could dose-dependently delay the VPB incubation period, decline the occurrence of VT, and increase the arrhythmia inhibition rate, indicating that they both had strong anti-arrhythmic activities. For example, mithaconitine could prolong the VPB incubation period in a dose-dependent manner, and the latency of VPB at 0.30 mg/kg and 0.40 mg/kg doses was significantly longer than that of lidocaine and propafenone (P < 0.05, Figure 4(a)). Furthermore, the arrhythmia inhibition rate Evidence-Based Complementary and Alternative Medicine of 0.40 mg/kg group was 50.0%, slightly higher than lidocaine (44.4%) (Figure 6(a)). Similar results were obtained for Δ 15(16) -16-demethoxyindaconitine. Δ 15(16) -16demethoxyindaconitine could dose-dependently delay the onset time of VPB, and the VPB incubation periods were significantly longer than that of propafenone and lidocaine at doses of 0.20 mg/kg, 0.30 mg/kg, and 0.40 mg/kg (P < 0.05, Figure 4(b)). e arrhythmia inhibition rate in 0.40 mg/kg group was 63.6%, higher than lidocaine (44.4%) ( Figure 6(b)), indicating that its anti-arrhythmic activity was also stronger than lidocaine. Besides, at the optimal dose of 0.40 mg/kg, the incubation periods of mithaconitine and Δ 15(16) -16-demethoxyindaconitine were 583.4 ± 219.7 s and 1211.0 ± 389.1 s, respectively. Namely, Δ 15(16) -16-demethoxyindaconitine exhibited a higher arrhythmia inhibition rate and a longer VPB incubation period than mithaconitine, rendering it more promising for development as anti-arrhythmic agents.

Conclusions
In summary, this study elucidated the structural transformation pathways of indaconitine during the sand frying process by comparing the structures of indaconitine and its transformed products. ere might be at least two transformation pathways that exist in the process of sand frying. Besides, cardiotoxicity assays and anti-arrhythmic activity experiments demonstrated that, compared with the prototype compound indaconitine, the two converted products exhibited lower cardiotoxicity and stronger anti-arrhythmic effects, which had the potential to be developed as antiarrhythmic drugs.

Data Availability
e data used to support the article are available in the article or supplementary materials.