- Email: [email protected]
Acute toxicity of berberine and its correlation with the blood concentration in mice
Food and Chemical Toxicology 48 (2010) 1105–1110
Contents lists available at ScienceDirect
Food and Chemical Toxicology journal homepage: www.elsevier.com/locate/foodchemtox
Acute toxicity of berberine and its correlation with the blood concentration in mice Michael M. Kheir a,b,c,1, Yugang Wang a,1, Lei Hua a, Jun Hu a, Lele Li a, Fan Lei a, Lijun Du a,* a
Laboratory of Pharmaceutical Science, School of Life Science, School of Medicine, Tsinghua University, Beijing 100084, China Department of Chemistry, University of Michigan, Ann Arbor, Michigan 48109, USA c Department of Psychology, University of Michigan, Ann Arbor, Michigan 48109, USA b
a r t i c l e
i n f o
Article history: Received 29 October 2009 Accepted 29 January 2010
Keywords: Berberine Acute toxicity HPLC Pharmacokinetics Mice
a b s t r a c t The aim of this study was to investigate the LD50 (median lethal dosage) of berberine (BBR) through three different routes of injection in mice: intravenous (IV) injection, intraperitoneal (IP) injection, and intragastric (IG) oral administration. The concentration of BBR in blood from their IG doses (10.4, 20.8, 41.6, and 83.2 g/kg) and the content relationship of BBR among different injections were analyzed by high-performance liquid chromatography (HPLC). The LD50 of BBR from IV and IP injections is 9.0386 and 57.6103 mg/kg, respectively; but no LD50 was found in the IG group. A significant difference in bioavailability was observed between the different routes. Furthermore, the concentration of BBR in the blood from different IG doses was also significantly different. However, we discovered an interesting phenomenon indicating that the absorption of BBR by oral administration has a limit, therefore, explaining the difficulty in obtaining an LD50 of BBR for IG injection. From the analysis of BBR content in blood after various administrations, we hypothesized that not only does the concentration of BBR in blood contribute to its acute toxicity, but also the routes of administration may be an important facet that affects this toxicity evaluation. Ó 2010 Elsevier Ltd. All rights reserved.
1. Introduction A natural product, berberine (BBR) is a quaternary ammonium salt from the group of isoquinoline alkaloids (Fig. 1). It is the main bioactive component in plants such as Berberis, goldenseal (Hydrastis canadensis), and Coptis chinensis. As a traditional medicine or dietary supplement in China, some prescriptions contain large concentrations of BBR because in conjunction with its long history of treatment in clinical practice, people have always believed that it has very low toxicity and superior effects in many diseases. According to modern research, it has been reported that BBR has some activity against and has already been used to treat dysentery, hypertension, inflammation, and liver disease in China and Japan (Chen et al., 2007). BBR has also been shown to exert antibiotic (Mirska et al., 1972; Wang et al., 2009), anti-inflammatory (Zhou and Mineshita, 2000), antineoplastic (Kuo et al., 1995), antiarrhythmic (Wang et al., 1994), antiproliferative (Ko et al., 2000), immunosuppressive (Marinova et al., 2000), anti-diabetic (Yin et al., 2008) and anti-hyperlipidemic functions (Zhao et al., 2008; Kong et al., 2008). Recent research has also included reports of neuroprotective properties (Zhou et al., 2008), antidepressive (Kulkarni and Dhir, 2008) and anticancer effects (Liu et al., 2009). BBR is absorbed poorly via oral administration whether administered alone or in combination with other drugs (Hua et al., 2007; Wang et al., 2005).
However, even though people believe that it is a safe compound, there is still no thorough report of the acute toxicity of such a commonly used drug. Besides the effects mentioned above, BBR needs to be studied to determine what concentration in the blood is safe. The acute toxicity of BBR via intravenous and intraperitoneal injections was found in order to provide comparison to the acute toxicity by means of oral administration. This research found the median lethal dosage (LD50) of BBR through two of the three different routes of injection in mice and the largest safe dose in oral administration. Then HPLC was used to explore the reason why different administration routes show different levels of toxicity. 2. Materials and methods 2.1. Experimental animals, drugs and chemicals Male and female ICR mice weighing 18–22 g were housed in temperature- and humidity-controlled rooms, kept on a 12 h light/dark cycle and provided with unrestricted amount of rodent chow and drinkable water. All procedures for animal experimentation were in accordance with the guidelines of China for animal care, which was conformed to the internationally accepted principles in the care and use of experimental animals. Berberine hydrochloride was obtained from the Beijing Ocean Pharmacy Co. Ltd. (Batch No. 090601). Berberine hydrochloride standards were purchased from the National Institute for the Control of Pharmaceutical and Biological Products (Batch No. 110713-200609). 2.2. Evaluation of berberine’s median lethal dosage by three routes of injection
* Corresponding author. Tel./fax: +86 10 62773630. E-mail address: [email protected] (L. Du). 1 Equal to contribution. 0278-6915/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.fct.2010.01.033
2.2.1. Experimental design and administration of berberine The methods to administer berberine to the mice included three routes of drug delivery: intravenous injection (IV), intraperitoneal injection (IP), and intragastric
1106
M.M. Kheir et al. / Food and Chemical Toxicology 48 (2010) 1105–1110
O O
N
H3CO
+
OCH3
ples was defined as the lowest concentration on the calibration curve for which assay imprecision (coefficient of variation, CV) was lower than 10% and was 10 times as much as LOD. 2.3.4. Recovery The recovery of BBR from the samples was evaluated using three different concentrations (0.0054, 0.134 and 3.36 lg/mL) covering the linear range of the standard curve. After the samples were processed according to the methods mentioned earlier, the resulting peak heights were compared to the BBR standard carried in mobile phase to provide the recovery values.
Fig. 1. The chemical structure of berberine. administration (IG). For IV and IP injection, 0.1 mL of berberine (BBR) was given for every 10 g of mouse body weight. For IG administration, 0.4 mL of BBR was given for every 10 g of mouse body weight. The method used to determine the pharmacological kinetics of BBR was an HPLC instrument. HPLC samples were made from collecting blood through the supraorbital vein of mice after certain times specific to the route of administration. 2.2.2. Median lethal dosage and the maximum tolerance dosage In determining the median lethal dosage (LD50), berberine was delivered via different routes and the mice were observed for 2 weeks. The mortality rate of the mice was recorded to, ultimately, calculate the LD50. Since the mice were tolerant to berberine by oral administration at a 10.4 g/kg dosage (maximum volume and concentration of BBR reagent), we altered the procedure for oral delivery to give multi-injections to increase the overall dosage of BBR, and hence the toxicity of BBR. We injected BBR once every hour (for accumulated dosages of 20.8, 41.6, and 83.2 g/kg, respectively) and recorded the number of dead mice in each dosage group. Meanwhile, another experiment was conducted at the same time as the multi-injection above, but we also obtain a blood sample through the supraorbital vein of these mice 1 h after BBR administration. The blood sample was prepared for the determination of BBR concentration as described below. 2.2.3. HPLC system The HPLC system consisted of a Waters 600 pump, a Rheodyne 7725i manual injector, and a 2487 UV–vis multi-wavelength detector (Waters, USA). A Zorbax SB-C18 reversed-phase column, 4.6 mm 150 mm i.d., 5 lm (Agilent, USA), was used. The signals from the detector were collected and analyzed with a computer equipped with a Waters Millennium32 Chromatography Manager. The mobile phase was water (0.2% triethylamine, pH 3.0 by phosphoric acid)–methanol (72:28, v/v), filtered through a 0.45 lm Millipore filter and degassed prior to use. The flow-rate was 1.0 mL/min. The injection volume was 20 lL. Detection was performed at a wavelength of 347 nm at room temperature (25–28 °C). 2.3. HPLC procedure 2.3.1. Preparation of plasma samples (Wang et al., 2005) Each collected blood sample was immediately transferred to a centrifuge tube and centrifuged at 12,000 rpm for 15 min. The supernatant was then extracted, transferred to a glass test tube, and mixed with methanol (plasma:methanol = 1:5) by vortex for 1 min, and ultrasound for 30 min. The resulting denatured protein precipitate was separated by centrifugation at 12,000 rpm for 15 min. The supernatant was collected and centrifuged again in centrifuge tubes to ensure that all proteins were removed from the solution. The supernatant was then put in a penicillin glass bottle. The test tubes were washed again with methanol and then centrifuged to obtain a higher yield of plasma recovery and the supernatant was collected again into their corresponding penicillin glass bottles. They were left to be evaporated at 37 °C and then the residue was dissolved in 100 lL of methanol. The supernatant was separated by another centrifugation at 12,000 rpm for 15 min. A 10 lL volume of each sample solution was injected into the HPLC system for analysis. Data from these samples were used to construct pharmacokinetic profiles by plotting drug concentration versus time. The same sample handling process was used for the determination of recovery and precision of BBR in plasma. 2.3.2. Calibration curve The BBR standard was dissolved in methanol and diluted to give concentrations of 0.0054, 0.0269, 0.134, 0.672 and 3.36 lg/mL. A calibration curve was constructed from the peak heights of BBR against the concentrations using un-weighted linear regression. The concentrations of BBR in the samples were determined using the regression parameters obtained from the calibration curve. Calibration standards were included in every analytical batch of samples. Ten microliter volumes of BBR standards were quantified by HPLC and a standard curve was derived. 2.3.3. Limit of detection and quantification The limit of detection (LOD) was determined as the lowest concentration that could be detected with acceptable accuracy and precision, which was achieved from the plot three times above the noise level. The limit of quantification (LOQ) in sam-
2.3.5. Precision and accuracy The precision of a quantitative method is the degree of agreement among the individual test results when the procedure is applied repeatedly to multiple samplings. It was measured by repeatedly injecting a ready-made sample pool and expressed as the relative standard deviation of the results. Analyses with three different concentrations (0.0054, 0.134 and 3.36 lg/mL) of BBR were performed. To determine the intra-day variance, the assays were carried out on the same samples at different times during 1 day. Inter-day variance was determined by assaying the spiked samples over three consecutive days at the same time each day. Coefficients of variation (CV) were calculated from these values. 2.4. Statistical analysis All values were expressed as the mean ± SD. The data were statistically analyzed by Excel software, and the t-test was carried out. P < 0.05 was accepted as statistical significance.
3. Results 3.1. LD50 and the maximum tolerance dosage The LD50 values that we determined for IV and IP routes of drug delivery were 9.0386 and 57.6103 mg/kg, respectively (Table 1). However, the LD50 could not be determined from IG experimental data although a 30% mortality rate was found among mice in the two highest dosage groups (41.6 and 83.2 g/kg) that orally administered the drug (Table 2). 3.2. Pharmacological kinetics of berberine 3.2.1. Calibration curve The calibration curve constructed for mice blood plasma was linear over the concentration range of 0.0054, 0.134, and 3.36 lg/mL. The correlation coefficient (r2) was 0.9997, suggesting a good linear relationship between peak areas and concentrations
Table 1 Summary of LD50 of different routes of injection with berberine and other statistical analysis.
Log(LD50) LD50 (mg/kg) Standard deviation of log(LD50) 95% Confidence interval for LD50
IV
IP
0.9561 9.0386 0.0196 (8.2737, 9.8742)
1.7605 57.6103 0.0874 (38.8329, 85.4673)
These data were calculated by the Software of LD50CAL2.0.
Table 2 Data of mice after oral administration of berberine at different dosages.
*
Dose (g/ kg)
Number of dead (1)
5.2 10.4 20.8 41.6 83.2
0 0 0 3 2
(0)* (0) (0) (1) (1)
Number of alive (2)
Total number (3)
Rate of death (1)/(3)
2 10 10 7 6
2 10 10 10 8
0 0 0 0.30 0.25
(1) (5) (5) (4) (3)
() Number of males is marked in parentheses.
(1) (5) (5) (5) (4)
M.M. Kheir et al. / Food and Chemical Toxicology 48 (2010) 1105–1110
1107
of BBR. The equation of the calibration curve was as follows: y = 439.64x + 1874.1, where x is the concentration of BBR in the blood and y is the area under the curve. The retention time of BBR in the system we used was 5.317 min, as seen in the chromatograms in (Fig. 2). 3.2.2. Recovery The loss of BBR due to the extraction process was determined by relating the data obtained by the direct injection of BBR standard dissolved in the mobile phase to the data obtained after the whole extraction procedure. To isolate BBR from the plasma, methanol was employed because of berberine’s ease in dissolving in it. The recovery values of 0.0054, 0.134 and 3.36 lg/mL from blood plasma were 88.79%, 88.01% and 85.23%, respectively. 3.2.3. Precision and accuracy The intra-day and inter-day precision, as well as the accuracy, were evaluated with three different concentrations. The coefficients of variation for intra-day precision of 0.0054, 0.134, and 3.36 lg/mL were 3.5%, 2.5% and 0.4%, respectively, while the coefficients of variation for inter-day precision of the same concentrations were 1.1%, 2.9% and 3.3%, respectively, indicating that the method was quite precise. 3.2.4. Limit of detection and quantification The limit of detection (LOD) was calculated as 0.00107 lg/mL using the equation, LOD = yB + 3SB (yB = blank signal, SB = standard deviation of the three peak areas of the minimum concentration of BBR) and the limit of quantification (LOQ) was 0.0054 lg/mL, which is sufficient for routine pharmacokinetic monitoring.
0.006
A
AU
0.004
Fig. 3. The time-course of berberine (BBR) blood concentrations after different injection routes. Each point and bar represents the mean ± SD. (A) The BBR blood concentration after intragastric oral administration at different times (n = 5; except for the group at 1 h, n = 9); (B) the BBR blood concentration after intraperitoneal administration at different times (n = 5; except for the group at 30 min, n = 8); (C) the BBR blood concentration after intravenous injection at different times (n = 5; except for the group at 5 min, n = 8).
0.002 0.000 -0.002 0.00
2.00
4.00
6.00
8.00
10.00
12.00
14.00
Minutes 0.20
B
5.317
AU
0.15 0.10 0.05 0.00 0.00
2.00
4.00
6.00
8.00
10.00
12.00
14.00
Minutes
0.030
C 5.375
AU
0.020 0.010 0.000 0.00
2.00
4.00
6.00
8.00
10.00
12.00
14.00
Minutes Fig. 2. Chromatograms for the determination of berberine (BBR) concentration in samples of mice blood plasma: (A) chromatogram of blank plasma; (B) chromatogram of BBR standard; (C) chromatogram of blood plasma drawn 60 min after oral administration of BBR (20.8 g/kg).
3.2.5. The time-course of BBR in blood after different routes of administration The LD50 dosages for IV and IP, and the dosages of 41.6 and 83.2 g/kg for IG injection were used to investigate the dynamic time-course of BBR in blood after each administration. The time at which the BBR blood concentration reached its peak after IV, IP and IG administration was at 5 min, 30 min and 1 h, respectively (Fig. 3).
3.2.6. The concentration limit of BBR in blood after oral administration in relation to death of mice In order to explain why the LD50 cannot be determined in the IG experiment, four different dosages of BBR (from a safe dosage to a dangerous dosage) in blood samples from orally administered mice were analyzed by HPLC. When the IG dose was 20.8 g/kg, the BBR concentration in blood plasma was 0.168 lg/mL and none of the mice had died. However, as the dose was increased to 41.6 g/kg, BBR blood concentration escalated to 0.432 lg/mL and the mice began to die at a mortality rate of approximately 30%. It is evident that the increase in dosage to 41.6 g/kg was accompanied with an increase in BBR concentration in the blood. However, as the dosage continues to increase up to 83.2 g/kg, the concentration of BBR in the blood is approximately the same as in the dosage of 41.6 g/kg. Moreover, the death rate does not change significantly; even up to a dosage of 83.2 g/kg where the mortality rate was only 25%. Therefore, this finding indicates that the BBR blood concentration of 0.432 lg/mL, corresponding to the dosage of 41.6 g/kg, is the
1108
M.M. Kheir et al. / Food and Chemical Toxicology 48 (2010) 1105–1110
Fig. 4. Mortality of mice and berberine (BBR) blood concentration after oral administration of BBR. The data represent the mean ± SEM values (in the 10.4 and 20.8 g/kg group, n = 4; in the 41.6 and 83.2 g/kg group, n = 9). The correlation coefficients (r) of dose-mortality and dose-blood concentration are 0.846820 and 0.920978, respectively. (*) p < 0.05, as compared to the blood concentration of 20.8 g/kg.
Fig. 5. Concentrations of berberine in blood after different injection routes. Intravenous (IV) administration with dosage of 9.0386 mg/kg (LD50). Intraperitoneal (IP) administration with dosage of 57.61 mg/kg (LD50). Each point and bar represents the mean ± SD (in the IV, IP, 41.6 g/kg oral administration and 83.2 g/kg oral administration group, n = 8; in the 10.4 and 20.8 g/kg oral administration group, n = 3). (**) p < 0.01, as compared to the IV group.
toxicity limit via oral administration. Furthermore, based on the experimental results, a suggested safe dosage is 20.8 g/kg (or a BBR blood concentration of 0.168 lg/mL) while a toxic dosage would be at 41.6 g/kg, which corresponds to a BBR blood concentration of 0.432 lg/mL (Fig. 4). 3.2.7. The bioavailability of BBR varies with different types of drug administration According to the LD50 data, the LD50 value in IP administration is about 6.5 times as much as IV injection and the rate of mortality in the IG experiment did not increase after the dosage of 41.6 g/kg. In order to explain such phenomena, HPLC was used to analyze blood samples from these different types of drug administrations. In order to find the bioavailability of BBR, the LD50 values for IV and IP were used (9.0386 mg/kg in IV and 57.61 mg/kg in IP); but for oral administration, dosages of 10.4, 20.8, 41.6, and 83.2 g BBR/kg mouse body weight were used. Blood was drawn at different time depending on the type of drug administration and the concentrations of BBR were determined by HPLC. The BBR blood concentration due to IV injection was the median lethal dose, which is equal to the BBR blood concentration due to IG injection at doses of 41.6 and 83.2 g/kg, the two dosages where mortality in mice is observed. In other words, the dosage of 41.6 g/kg by oral administration is the toxic dose (Fig. 5). Regarding the BBR blood concentration via IV injection as 100%, the bioavailability of the drug deliveries is 100%, 20.946%, 0.0001%, 0.002%, 0.024% and 0.011% for blood drawn 5 min after IV, 30 min after IP, and 60 min after IG administration at four successively increasing dosages, respectively. It is clear that the bioavailability of BBR by oral administration is extremely lower than that by IV and IP administrations (Fig. 6). 4. Discussion The LD50 of BBR was investigated via three different routes of administration. In order to evaluate the acute toxicity data accurately, after determining the lowest dosage that would kill a mouse and the highest dosage that did not cause mortality in mice, the following equation was used to determine the two dosages (arbitrarily denoted as D1 and D2) to be used in between to find the LD50:
1=K ¼ ðn1Þ
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi Dmax =Dmin
ð1Þ
where Dmax is the highest concentration used (the lowest concentration capable of mortality in all mice) and Dmin is the lowest concentration used (the highest concentration that is safe, i.e., no
Fig. 6. Ratios of the berberine concentration in blood of different injection routes to the IV injection group. Intravenous (IV) administration with dosage of 9.0386 mg/kg (LD50). Intraperitoneal (IP) administration with dosage of 57.61 mg/kg (LD50). The data represent the mean values (in the IV, IP, 41.6 g/kg oral administration and 83.2 g/kg oral administration group, n = 8; in the 10.4 and 20.8 g/kg oral administration group, n = 3).
mortality in any mice). The ratio used is 1:K:K2:K3 = Dmin:D1:D2: Dmax. N is the number of total dosages used to determine the LD50, which, in this case, was 4 and is the same for each type of administration route. This equation is based on the sequential method of determining the median lethal dosage (Chanter and Heywood, 1982). The results indicated that the LD50 of BBR via IV and IP injections is 9.0386 ± 0.8003 mg/kg and 57.6103 ± 23.3172 mg/kg, respectively, which is consistent with a previous report of the LD50 being 50 mg/kg by IP administration (Anis et al., 1999). However, referring to the previous reports, no LD50 of BBR in oral administration was found at the dosage of 10.4 g/kg. In order to explore the reason why different drug administration routes with the same compound show different acute toxicities, HPLC was used to analyze the BBR blood concentration after injection via the three different routes. The time-course of BBR in blood after each route was foremost investigated to find the proper time of sampling. The concentration of BBR in blood after IP (LD50 dosage) and IG (41.6 g/kg dosage) administration reaches the highest peak after 30 min and 1 h, correspondingly. For the IV group, the graph peaked at the first time-point we obtained, which was at 5 min where there was the highest blood concentration of BBR; 5 min was the shortest time interval between injection and sampling with which we could make sure that every sampling was under the same condition. The blood concentration of BBR in samples from IG at dosage of 83.2 g/kg remained the highest for the first 2 h; we believe that this phenomenon concerns the clear-up process carried out by animals. Furthermore, this phenomenon can be explained by the length of time for the administration of a dosage that is high (7 h) and the presence of too much drug to be cleared up and absorbed by the intestinal system at one time. According to the experimental data, the bioavailability of BBR decreases in the following order of administration routes: IV >
M.M. Kheir et al. / Food and Chemical Toxicology 48 (2010) 1105–1110
IP > IG, which corresponds with the increasing LD50 value of BBR in these three routes. We hypothesized that the bioavailability could be one of the most important elements that affect the LD50 value and that the concentration of BBR in the blood could be an explanation for the various acute toxicities of the different routes. To prove this hypothesis, 4 groups of mice were given BBR at different dosages (10.4, 20.8, 41.6, and 83.2 g/kg) via intragastric oral administration and blood samples were collected 1 h later to be analyzed by HPLC. From the data, we determined that the blood concentration of BBR after oral administration forms an ‘‘S”-like curve (Fig. 4). With an increase in dosage, an accompanying increase in blood concentration of BBR was found until a point where the dosage had reached a high enough level that the concentration of BBR in the blood reaches a limit in oral administration. This may be the reason why the mortality of mice at a dosage of 41.6 g/kg was not significantly different from the dosage of 83.2 g/kg (Fig. 5). It is also suggested that at higher doses than 41.6 g/kg, the blood concentration of berberine maintains a steady-state concentration. Surprisingly, however, the BBR blood concentration in the mice that died via the oral route of injection is nearly the same as the LD50 dosage via IV administration, and hence nearly the same as the BBR blood concentration via IV because of its 100% bioavailability. Using the BBR blood concentration from the IV group as a standard, there was no statistical significant difference between the highest peaks of each route. From this observation, we observe that the acute toxic reactions of BBR occur at a certain concentration of BBR in blood (about 0.38 lg/mL), despite the route taken to enter the bloodstream (Fig. 5). The blood concentration of BBR after oral administration with a safe dosage was only 0.043 times (0.017/0.395 lg/mL) lower than the concentration via IV injection with its LD50 dosage. However, the dosage of oral administration was more than about 1105 times (20.8/0.00904 g/kg) higher than the dosage of IV injection. This interesting phenomenon may demonstrate that the acute toxicity of a certain chemical compound not only correlates with its concentration in the blood after administration, but also relates to the type of administration route used to introduce the drug into the blood. This result also suggests that the absorption of BBR by the animal’s intestine system has its own limit, and no matter how much the orally administered dosage has increased, the absorption rate would not increase at this internal limit; the excess berberine will be excreted or removed by increasing metabolism (Zuo et al., 2006). In our results, we found that 20.8 g BBR/kg of body weight is safe for oral administration in mice. The mice also have a metabolic rate per kg body weight that is approximately (and conservatively) seven times higher than humans (Terpstra, 2001). Therefore, BBR enters the bloodstream much more quickly in mice. In light of this, the safety dosage for humans would be approximately 1/7 of 20.81 g BBR/kg (2.97 g BBR/kg of human body weight). We can reasonably infer from these data that the clinically recommended dosage of 15 mg BBR/kg human body weight (three 100 mg berberine tablets taken three times a day for the average 70 kg adult human) is safe. In the article, ‘‘Berberine is a novel cholesterol-lowering drug working through a unique mechanism distinct from statins” (Kong et al., 2004), we can see that the dosage used in the experiment by Kong and colleagues was safe. In the article, it states that the human participants were administered 0.5 g of BBR twice daily. Assuming the average human participant weighed on approximately 70 kg and that the 500 mg of BBR was given every 12 h, it is clearly well below the safety dosage of BBR. 5. Conclusion This is the first report that involves a thorough and well-conducted study of berberine’s acute toxicity via different injection
1109
routes and their corresponding blood concentrations systematically. The LD50 of BBR for IV and IP injections was determined for the first time in mice. Attempting to find the LD50 of BBR for IG administration proved to be strenuous and the LD50 could not be established, but the limitation of BBR blood concentration and its effect on mice mortality rate was explored. This study will be of great benefit to the understanding and practice of BBR use in vivo and in vitro.
Conflict of interest The authors declare that there are no conflicts of interest. Acknowledgment This project (90713043, 30801523, and 30973896) was supported by the National Natural Science Foundation of China. References Anis, K.V., Kuttan, G., Kuttan, R., 1999. Role of berberine as an adjuvant response modifier during tumour therapy in mice. Pharmaceutical Pharmacology Communications 5, 697–700. Chanter, D.O., Heywood, R., 1982. The LD50 test: some considerations of precision. Toxicology Letters 10 (2–3), 303–307. Chen, Y.Y., Wang, X.L., Sun, H., Xing, D.M., Hu, J., Wai, Z.H., Du, L.J., 2007. Characterization of the transportation of berberine in Coptidis rhizoma extract through rat primary cultured cortical neurons. Biomedical Chromatograph 22, 28–33. Hua, W.Y., Ding, L., Chen, Y., Gong, B., He, J.C., Xu, G.L., 2007. Determination of berberine in human plasma by liquid chromatography–electrospray ionization– mass spectrometry. Journal of Pharmaceutical and Biomedical Analysis 44, 931– 937. Ko, W.H., Yao, X.Q., Lau, C.W., Law, W.I., Chen, Z.Y., Kwok, W., Ho, K., Huang, Y., 2000. Vasorelaxant and antiproliferative effects of berberine. European Journal of Pharmacology 399, 187–196. Kong, W.J., Wei, J., Abidi, P., Lin, M., Inaba, S., Li, C., Wang, Y.L., Wang, Z.H., Si, S.Y., Pan, H.N., Wang, S.K., Wu, J.D., Wang, Y., Li, Z.R., Liu, J.W., Jiang, J.D., 2004. Berberine is a novel cholesterol-lowering drug working through a unique mechanism distinct from statins. Nature Medicine 10 (12), 1344–1351. Kong, W.J., Wei, J., Zuo, Z.Y., Wang, Y.M., Song, D.Q., You, X.F., Zhao, L.X., Pan, H.N., Jiang, J.D., 2008. Combination of simvastatin with berberine improves the lipid-lowering efficacy. Metabolism Clinical and Experimental 57, 1029– 1037. Kulkarni, S.K., Dhir, A., 2008. On the mechanism of antidepressant-like action of berberine chloride. European Journal of Pharmacology 589, 163–172. Kuo, C.L., Chou, C.C., Benjamin, Y.M., 1995. Berberine complexes with DNA in the berberine-induced apoptosis in human leukemic HL-60 cells. Cancer Letters 93, 193–200. Liu, J., He, C., Zhou, K., Wang, J., Kang, J.X., 2009. Coptis extracts enhance the anticancer effect of estrogen receptor antagonists on human breast cancer cells. Biochemical and Biophysical Research Communications 378 (2), 174– 178. Marinova, E.K., Nikolova, D.B., Popova, D.N., Gallacher, G.B., Ivanovska, N.D., 2000. Suppression of experimental autoimmune tubulointerstitial nephritis in BALB/C mice by berberine. Immunopharmacology 48, 9–16. Mirska, I., Kedzia, H., Kowalewski, Z., Kedzia, W., 1972. The effect of berberine sulfate on healthy mice infected with Candida albicans. Archive of Immunology Therapeutics Experiment 20, 921–929. Terpstra, A.H.M., 2001. Differences between humans and mice in efficacy of the body fat lowering effect of conjugated linoleic acid: role of metabolic rate. Journal of Nutrition 131, 2067–2068. Wang, Y.X., Yao, X.J., Tan, Y.H., 1994. Effects of berberine on delayed after depolarizations in ventricular muscles in vitro and in vivo. Journal of Cardiovascular Pharmacology 23, 716–722. Wang, X.L., Wang, R.F., Xing, D.M., Su, H., Ma, C., Ding, Y., Du, L.J., 2005. Kinetic difference of berberine between hippocampus and plasma in rat after intravenous administration of Coptidis rhizoma extract. Life Sciences 77, 3058–3067. Wang, X.Q., Qiu, S.J., Yao, X., Tang, T.T., Dai, K.R., Zhu, Z.A., 2009. Berberine inhibits staphylococcus epidermidis adhesion and biofilm formation on the surface of titanium alloy. Journal of Orthopaedic Research 27, 1487– 1492. Yin, J., Xing, H.L., Ye, J.P., 2008. Efficacy of berberine in patients with type 2 diabetes mellitus. Metabolism Clinical and Experimental 57, 712–717. Zhao, W., Zhou, Z.X., Kong, W.J., Jiang, J.D., 2008. Reduction of blood lipid by berberine in hyperlipidemic patients with chronic hepatitis or liver cirrhosis. Biomedicine and Pharmacotherapy 62, 730–731.
1110
M.M. Kheir et al. / Food and Chemical Toxicology 48 (2010) 1105–1110
Zhou, H., Mineshita, S., 2000. The effect of berberine chloride on experimental colitis in rats in vivo and in vitro. Journal of Pharmacology and Experimental Therapeutics 294, 822–829. Zhou, X.Q., Zeng, X.N., Kong, H., Sun, X.L., 2008. Neuroprotective effects of berberine on stroke models in vitro and in vivo. Neuroscience Letters 447, 31–36.
Zuo, F., Nakamura, N., Akao, T., Hattori, M., 2006. Pharmacokinetics of berberine and its main metabolites in conventional and pseudo germ-free rats determined by liquid chromatography/ion trap mass spectrometry. Drug Metabolism and Disposition 34 (12), 2064–2072.
Contents lists available at ScienceDirect
Food and Chemical Toxicology journal homepage: www.elsevier.com/locate/foodchemtox
Acute toxicity of berberine and its correlation with the blood concentration in mice Michael M. Kheir a,b,c,1, Yugang Wang a,1, Lei Hua a, Jun Hu a, Lele Li a, Fan Lei a, Lijun Du a,* a
Laboratory of Pharmaceutical Science, School of Life Science, School of Medicine, Tsinghua University, Beijing 100084, China Department of Chemistry, University of Michigan, Ann Arbor, Michigan 48109, USA c Department of Psychology, University of Michigan, Ann Arbor, Michigan 48109, USA b
a r t i c l e
i n f o
Article history: Received 29 October 2009 Accepted 29 January 2010
Keywords: Berberine Acute toxicity HPLC Pharmacokinetics Mice
a b s t r a c t The aim of this study was to investigate the LD50 (median lethal dosage) of berberine (BBR) through three different routes of injection in mice: intravenous (IV) injection, intraperitoneal (IP) injection, and intragastric (IG) oral administration. The concentration of BBR in blood from their IG doses (10.4, 20.8, 41.6, and 83.2 g/kg) and the content relationship of BBR among different injections were analyzed by high-performance liquid chromatography (HPLC). The LD50 of BBR from IV and IP injections is 9.0386 and 57.6103 mg/kg, respectively; but no LD50 was found in the IG group. A significant difference in bioavailability was observed between the different routes. Furthermore, the concentration of BBR in the blood from different IG doses was also significantly different. However, we discovered an interesting phenomenon indicating that the absorption of BBR by oral administration has a limit, therefore, explaining the difficulty in obtaining an LD50 of BBR for IG injection. From the analysis of BBR content in blood after various administrations, we hypothesized that not only does the concentration of BBR in blood contribute to its acute toxicity, but also the routes of administration may be an important facet that affects this toxicity evaluation. Ó 2010 Elsevier Ltd. All rights reserved.
1. Introduction A natural product, berberine (BBR) is a quaternary ammonium salt from the group of isoquinoline alkaloids (Fig. 1). It is the main bioactive component in plants such as Berberis, goldenseal (Hydrastis canadensis), and Coptis chinensis. As a traditional medicine or dietary supplement in China, some prescriptions contain large concentrations of BBR because in conjunction with its long history of treatment in clinical practice, people have always believed that it has very low toxicity and superior effects in many diseases. According to modern research, it has been reported that BBR has some activity against and has already been used to treat dysentery, hypertension, inflammation, and liver disease in China and Japan (Chen et al., 2007). BBR has also been shown to exert antibiotic (Mirska et al., 1972; Wang et al., 2009), anti-inflammatory (Zhou and Mineshita, 2000), antineoplastic (Kuo et al., 1995), antiarrhythmic (Wang et al., 1994), antiproliferative (Ko et al., 2000), immunosuppressive (Marinova et al., 2000), anti-diabetic (Yin et al., 2008) and anti-hyperlipidemic functions (Zhao et al., 2008; Kong et al., 2008). Recent research has also included reports of neuroprotective properties (Zhou et al., 2008), antidepressive (Kulkarni and Dhir, 2008) and anticancer effects (Liu et al., 2009). BBR is absorbed poorly via oral administration whether administered alone or in combination with other drugs (Hua et al., 2007; Wang et al., 2005).
However, even though people believe that it is a safe compound, there is still no thorough report of the acute toxicity of such a commonly used drug. Besides the effects mentioned above, BBR needs to be studied to determine what concentration in the blood is safe. The acute toxicity of BBR via intravenous and intraperitoneal injections was found in order to provide comparison to the acute toxicity by means of oral administration. This research found the median lethal dosage (LD50) of BBR through two of the three different routes of injection in mice and the largest safe dose in oral administration. Then HPLC was used to explore the reason why different administration routes show different levels of toxicity. 2. Materials and methods 2.1. Experimental animals, drugs and chemicals Male and female ICR mice weighing 18–22 g were housed in temperature- and humidity-controlled rooms, kept on a 12 h light/dark cycle and provided with unrestricted amount of rodent chow and drinkable water. All procedures for animal experimentation were in accordance with the guidelines of China for animal care, which was conformed to the internationally accepted principles in the care and use of experimental animals. Berberine hydrochloride was obtained from the Beijing Ocean Pharmacy Co. Ltd. (Batch No. 090601). Berberine hydrochloride standards were purchased from the National Institute for the Control of Pharmaceutical and Biological Products (Batch No. 110713-200609). 2.2. Evaluation of berberine’s median lethal dosage by three routes of injection
* Corresponding author. Tel./fax: +86 10 62773630. E-mail address: [email protected] (L. Du). 1 Equal to contribution. 0278-6915/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.fct.2010.01.033
2.2.1. Experimental design and administration of berberine The methods to administer berberine to the mice included three routes of drug delivery: intravenous injection (IV), intraperitoneal injection (IP), and intragastric
1106
M.M. Kheir et al. / Food and Chemical Toxicology 48 (2010) 1105–1110
O O
N
H3CO
+
OCH3
ples was defined as the lowest concentration on the calibration curve for which assay imprecision (coefficient of variation, CV) was lower than 10% and was 10 times as much as LOD. 2.3.4. Recovery The recovery of BBR from the samples was evaluated using three different concentrations (0.0054, 0.134 and 3.36 lg/mL) covering the linear range of the standard curve. After the samples were processed according to the methods mentioned earlier, the resulting peak heights were compared to the BBR standard carried in mobile phase to provide the recovery values.
Fig. 1. The chemical structure of berberine. administration (IG). For IV and IP injection, 0.1 mL of berberine (BBR) was given for every 10 g of mouse body weight. For IG administration, 0.4 mL of BBR was given for every 10 g of mouse body weight. The method used to determine the pharmacological kinetics of BBR was an HPLC instrument. HPLC samples were made from collecting blood through the supraorbital vein of mice after certain times specific to the route of administration. 2.2.2. Median lethal dosage and the maximum tolerance dosage In determining the median lethal dosage (LD50), berberine was delivered via different routes and the mice were observed for 2 weeks. The mortality rate of the mice was recorded to, ultimately, calculate the LD50. Since the mice were tolerant to berberine by oral administration at a 10.4 g/kg dosage (maximum volume and concentration of BBR reagent), we altered the procedure for oral delivery to give multi-injections to increase the overall dosage of BBR, and hence the toxicity of BBR. We injected BBR once every hour (for accumulated dosages of 20.8, 41.6, and 83.2 g/kg, respectively) and recorded the number of dead mice in each dosage group. Meanwhile, another experiment was conducted at the same time as the multi-injection above, but we also obtain a blood sample through the supraorbital vein of these mice 1 h after BBR administration. The blood sample was prepared for the determination of BBR concentration as described below. 2.2.3. HPLC system The HPLC system consisted of a Waters 600 pump, a Rheodyne 7725i manual injector, and a 2487 UV–vis multi-wavelength detector (Waters, USA). A Zorbax SB-C18 reversed-phase column, 4.6 mm 150 mm i.d., 5 lm (Agilent, USA), was used. The signals from the detector were collected and analyzed with a computer equipped with a Waters Millennium32 Chromatography Manager. The mobile phase was water (0.2% triethylamine, pH 3.0 by phosphoric acid)–methanol (72:28, v/v), filtered through a 0.45 lm Millipore filter and degassed prior to use. The flow-rate was 1.0 mL/min. The injection volume was 20 lL. Detection was performed at a wavelength of 347 nm at room temperature (25–28 °C). 2.3. HPLC procedure 2.3.1. Preparation of plasma samples (Wang et al., 2005) Each collected blood sample was immediately transferred to a centrifuge tube and centrifuged at 12,000 rpm for 15 min. The supernatant was then extracted, transferred to a glass test tube, and mixed with methanol (plasma:methanol = 1:5) by vortex for 1 min, and ultrasound for 30 min. The resulting denatured protein precipitate was separated by centrifugation at 12,000 rpm for 15 min. The supernatant was collected and centrifuged again in centrifuge tubes to ensure that all proteins were removed from the solution. The supernatant was then put in a penicillin glass bottle. The test tubes were washed again with methanol and then centrifuged to obtain a higher yield of plasma recovery and the supernatant was collected again into their corresponding penicillin glass bottles. They were left to be evaporated at 37 °C and then the residue was dissolved in 100 lL of methanol. The supernatant was separated by another centrifugation at 12,000 rpm for 15 min. A 10 lL volume of each sample solution was injected into the HPLC system for analysis. Data from these samples were used to construct pharmacokinetic profiles by plotting drug concentration versus time. The same sample handling process was used for the determination of recovery and precision of BBR in plasma. 2.3.2. Calibration curve The BBR standard was dissolved in methanol and diluted to give concentrations of 0.0054, 0.0269, 0.134, 0.672 and 3.36 lg/mL. A calibration curve was constructed from the peak heights of BBR against the concentrations using un-weighted linear regression. The concentrations of BBR in the samples were determined using the regression parameters obtained from the calibration curve. Calibration standards were included in every analytical batch of samples. Ten microliter volumes of BBR standards were quantified by HPLC and a standard curve was derived. 2.3.3. Limit of detection and quantification The limit of detection (LOD) was determined as the lowest concentration that could be detected with acceptable accuracy and precision, which was achieved from the plot three times above the noise level. The limit of quantification (LOQ) in sam-
2.3.5. Precision and accuracy The precision of a quantitative method is the degree of agreement among the individual test results when the procedure is applied repeatedly to multiple samplings. It was measured by repeatedly injecting a ready-made sample pool and expressed as the relative standard deviation of the results. Analyses with three different concentrations (0.0054, 0.134 and 3.36 lg/mL) of BBR were performed. To determine the intra-day variance, the assays were carried out on the same samples at different times during 1 day. Inter-day variance was determined by assaying the spiked samples over three consecutive days at the same time each day. Coefficients of variation (CV) were calculated from these values. 2.4. Statistical analysis All values were expressed as the mean ± SD. The data were statistically analyzed by Excel software, and the t-test was carried out. P < 0.05 was accepted as statistical significance.
3. Results 3.1. LD50 and the maximum tolerance dosage The LD50 values that we determined for IV and IP routes of drug delivery were 9.0386 and 57.6103 mg/kg, respectively (Table 1). However, the LD50 could not be determined from IG experimental data although a 30% mortality rate was found among mice in the two highest dosage groups (41.6 and 83.2 g/kg) that orally administered the drug (Table 2). 3.2. Pharmacological kinetics of berberine 3.2.1. Calibration curve The calibration curve constructed for mice blood plasma was linear over the concentration range of 0.0054, 0.134, and 3.36 lg/mL. The correlation coefficient (r2) was 0.9997, suggesting a good linear relationship between peak areas and concentrations
Table 1 Summary of LD50 of different routes of injection with berberine and other statistical analysis.
Log(LD50) LD50 (mg/kg) Standard deviation of log(LD50) 95% Confidence interval for LD50
IV
IP
0.9561 9.0386 0.0196 (8.2737, 9.8742)
1.7605 57.6103 0.0874 (38.8329, 85.4673)
These data were calculated by the Software of LD50CAL2.0.
Table 2 Data of mice after oral administration of berberine at different dosages.
*
Dose (g/ kg)
Number of dead (1)
5.2 10.4 20.8 41.6 83.2
0 0 0 3 2
(0)* (0) (0) (1) (1)
Number of alive (2)
Total number (3)
Rate of death (1)/(3)
2 10 10 7 6
2 10 10 10 8
0 0 0 0.30 0.25
(1) (5) (5) (4) (3)
() Number of males is marked in parentheses.
(1) (5) (5) (5) (4)
M.M. Kheir et al. / Food and Chemical Toxicology 48 (2010) 1105–1110
1107
of BBR. The equation of the calibration curve was as follows: y = 439.64x + 1874.1, where x is the concentration of BBR in the blood and y is the area under the curve. The retention time of BBR in the system we used was 5.317 min, as seen in the chromatograms in (Fig. 2). 3.2.2. Recovery The loss of BBR due to the extraction process was determined by relating the data obtained by the direct injection of BBR standard dissolved in the mobile phase to the data obtained after the whole extraction procedure. To isolate BBR from the plasma, methanol was employed because of berberine’s ease in dissolving in it. The recovery values of 0.0054, 0.134 and 3.36 lg/mL from blood plasma were 88.79%, 88.01% and 85.23%, respectively. 3.2.3. Precision and accuracy The intra-day and inter-day precision, as well as the accuracy, were evaluated with three different concentrations. The coefficients of variation for intra-day precision of 0.0054, 0.134, and 3.36 lg/mL were 3.5%, 2.5% and 0.4%, respectively, while the coefficients of variation for inter-day precision of the same concentrations were 1.1%, 2.9% and 3.3%, respectively, indicating that the method was quite precise. 3.2.4. Limit of detection and quantification The limit of detection (LOD) was calculated as 0.00107 lg/mL using the equation, LOD = yB + 3SB (yB = blank signal, SB = standard deviation of the three peak areas of the minimum concentration of BBR) and the limit of quantification (LOQ) was 0.0054 lg/mL, which is sufficient for routine pharmacokinetic monitoring.
0.006
A
AU
0.004
Fig. 3. The time-course of berberine (BBR) blood concentrations after different injection routes. Each point and bar represents the mean ± SD. (A) The BBR blood concentration after intragastric oral administration at different times (n = 5; except for the group at 1 h, n = 9); (B) the BBR blood concentration after intraperitoneal administration at different times (n = 5; except for the group at 30 min, n = 8); (C) the BBR blood concentration after intravenous injection at different times (n = 5; except for the group at 5 min, n = 8).
0.002 0.000 -0.002 0.00
2.00
4.00
6.00
8.00
10.00
12.00
14.00
Minutes 0.20
B
5.317
AU
0.15 0.10 0.05 0.00 0.00
2.00
4.00
6.00
8.00
10.00
12.00
14.00
Minutes
0.030
C 5.375
AU
0.020 0.010 0.000 0.00
2.00
4.00
6.00
8.00
10.00
12.00
14.00
Minutes Fig. 2. Chromatograms for the determination of berberine (BBR) concentration in samples of mice blood plasma: (A) chromatogram of blank plasma; (B) chromatogram of BBR standard; (C) chromatogram of blood plasma drawn 60 min after oral administration of BBR (20.8 g/kg).
3.2.5. The time-course of BBR in blood after different routes of administration The LD50 dosages for IV and IP, and the dosages of 41.6 and 83.2 g/kg for IG injection were used to investigate the dynamic time-course of BBR in blood after each administration. The time at which the BBR blood concentration reached its peak after IV, IP and IG administration was at 5 min, 30 min and 1 h, respectively (Fig. 3).
3.2.6. The concentration limit of BBR in blood after oral administration in relation to death of mice In order to explain why the LD50 cannot be determined in the IG experiment, four different dosages of BBR (from a safe dosage to a dangerous dosage) in blood samples from orally administered mice were analyzed by HPLC. When the IG dose was 20.8 g/kg, the BBR concentration in blood plasma was 0.168 lg/mL and none of the mice had died. However, as the dose was increased to 41.6 g/kg, BBR blood concentration escalated to 0.432 lg/mL and the mice began to die at a mortality rate of approximately 30%. It is evident that the increase in dosage to 41.6 g/kg was accompanied with an increase in BBR concentration in the blood. However, as the dosage continues to increase up to 83.2 g/kg, the concentration of BBR in the blood is approximately the same as in the dosage of 41.6 g/kg. Moreover, the death rate does not change significantly; even up to a dosage of 83.2 g/kg where the mortality rate was only 25%. Therefore, this finding indicates that the BBR blood concentration of 0.432 lg/mL, corresponding to the dosage of 41.6 g/kg, is the
1108
M.M. Kheir et al. / Food and Chemical Toxicology 48 (2010) 1105–1110
Fig. 4. Mortality of mice and berberine (BBR) blood concentration after oral administration of BBR. The data represent the mean ± SEM values (in the 10.4 and 20.8 g/kg group, n = 4; in the 41.6 and 83.2 g/kg group, n = 9). The correlation coefficients (r) of dose-mortality and dose-blood concentration are 0.846820 and 0.920978, respectively. (*) p < 0.05, as compared to the blood concentration of 20.8 g/kg.
Fig. 5. Concentrations of berberine in blood after different injection routes. Intravenous (IV) administration with dosage of 9.0386 mg/kg (LD50). Intraperitoneal (IP) administration with dosage of 57.61 mg/kg (LD50). Each point and bar represents the mean ± SD (in the IV, IP, 41.6 g/kg oral administration and 83.2 g/kg oral administration group, n = 8; in the 10.4 and 20.8 g/kg oral administration group, n = 3). (**) p < 0.01, as compared to the IV group.
toxicity limit via oral administration. Furthermore, based on the experimental results, a suggested safe dosage is 20.8 g/kg (or a BBR blood concentration of 0.168 lg/mL) while a toxic dosage would be at 41.6 g/kg, which corresponds to a BBR blood concentration of 0.432 lg/mL (Fig. 4). 3.2.7. The bioavailability of BBR varies with different types of drug administration According to the LD50 data, the LD50 value in IP administration is about 6.5 times as much as IV injection and the rate of mortality in the IG experiment did not increase after the dosage of 41.6 g/kg. In order to explain such phenomena, HPLC was used to analyze blood samples from these different types of drug administrations. In order to find the bioavailability of BBR, the LD50 values for IV and IP were used (9.0386 mg/kg in IV and 57.61 mg/kg in IP); but for oral administration, dosages of 10.4, 20.8, 41.6, and 83.2 g BBR/kg mouse body weight were used. Blood was drawn at different time depending on the type of drug administration and the concentrations of BBR were determined by HPLC. The BBR blood concentration due to IV injection was the median lethal dose, which is equal to the BBR blood concentration due to IG injection at doses of 41.6 and 83.2 g/kg, the two dosages where mortality in mice is observed. In other words, the dosage of 41.6 g/kg by oral administration is the toxic dose (Fig. 5). Regarding the BBR blood concentration via IV injection as 100%, the bioavailability of the drug deliveries is 100%, 20.946%, 0.0001%, 0.002%, 0.024% and 0.011% for blood drawn 5 min after IV, 30 min after IP, and 60 min after IG administration at four successively increasing dosages, respectively. It is clear that the bioavailability of BBR by oral administration is extremely lower than that by IV and IP administrations (Fig. 6). 4. Discussion The LD50 of BBR was investigated via three different routes of administration. In order to evaluate the acute toxicity data accurately, after determining the lowest dosage that would kill a mouse and the highest dosage that did not cause mortality in mice, the following equation was used to determine the two dosages (arbitrarily denoted as D1 and D2) to be used in between to find the LD50:
1=K ¼ ðn1Þ
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi Dmax =Dmin
ð1Þ
where Dmax is the highest concentration used (the lowest concentration capable of mortality in all mice) and Dmin is the lowest concentration used (the highest concentration that is safe, i.e., no
Fig. 6. Ratios of the berberine concentration in blood of different injection routes to the IV injection group. Intravenous (IV) administration with dosage of 9.0386 mg/kg (LD50). Intraperitoneal (IP) administration with dosage of 57.61 mg/kg (LD50). The data represent the mean values (in the IV, IP, 41.6 g/kg oral administration and 83.2 g/kg oral administration group, n = 8; in the 10.4 and 20.8 g/kg oral administration group, n = 3).
mortality in any mice). The ratio used is 1:K:K2:K3 = Dmin:D1:D2: Dmax. N is the number of total dosages used to determine the LD50, which, in this case, was 4 and is the same for each type of administration route. This equation is based on the sequential method of determining the median lethal dosage (Chanter and Heywood, 1982). The results indicated that the LD50 of BBR via IV and IP injections is 9.0386 ± 0.8003 mg/kg and 57.6103 ± 23.3172 mg/kg, respectively, which is consistent with a previous report of the LD50 being 50 mg/kg by IP administration (Anis et al., 1999). However, referring to the previous reports, no LD50 of BBR in oral administration was found at the dosage of 10.4 g/kg. In order to explore the reason why different drug administration routes with the same compound show different acute toxicities, HPLC was used to analyze the BBR blood concentration after injection via the three different routes. The time-course of BBR in blood after each route was foremost investigated to find the proper time of sampling. The concentration of BBR in blood after IP (LD50 dosage) and IG (41.6 g/kg dosage) administration reaches the highest peak after 30 min and 1 h, correspondingly. For the IV group, the graph peaked at the first time-point we obtained, which was at 5 min where there was the highest blood concentration of BBR; 5 min was the shortest time interval between injection and sampling with which we could make sure that every sampling was under the same condition. The blood concentration of BBR in samples from IG at dosage of 83.2 g/kg remained the highest for the first 2 h; we believe that this phenomenon concerns the clear-up process carried out by animals. Furthermore, this phenomenon can be explained by the length of time for the administration of a dosage that is high (7 h) and the presence of too much drug to be cleared up and absorbed by the intestinal system at one time. According to the experimental data, the bioavailability of BBR decreases in the following order of administration routes: IV >
M.M. Kheir et al. / Food and Chemical Toxicology 48 (2010) 1105–1110
IP > IG, which corresponds with the increasing LD50 value of BBR in these three routes. We hypothesized that the bioavailability could be one of the most important elements that affect the LD50 value and that the concentration of BBR in the blood could be an explanation for the various acute toxicities of the different routes. To prove this hypothesis, 4 groups of mice were given BBR at different dosages (10.4, 20.8, 41.6, and 83.2 g/kg) via intragastric oral administration and blood samples were collected 1 h later to be analyzed by HPLC. From the data, we determined that the blood concentration of BBR after oral administration forms an ‘‘S”-like curve (Fig. 4). With an increase in dosage, an accompanying increase in blood concentration of BBR was found until a point where the dosage had reached a high enough level that the concentration of BBR in the blood reaches a limit in oral administration. This may be the reason why the mortality of mice at a dosage of 41.6 g/kg was not significantly different from the dosage of 83.2 g/kg (Fig. 5). It is also suggested that at higher doses than 41.6 g/kg, the blood concentration of berberine maintains a steady-state concentration. Surprisingly, however, the BBR blood concentration in the mice that died via the oral route of injection is nearly the same as the LD50 dosage via IV administration, and hence nearly the same as the BBR blood concentration via IV because of its 100% bioavailability. Using the BBR blood concentration from the IV group as a standard, there was no statistical significant difference between the highest peaks of each route. From this observation, we observe that the acute toxic reactions of BBR occur at a certain concentration of BBR in blood (about 0.38 lg/mL), despite the route taken to enter the bloodstream (Fig. 5). The blood concentration of BBR after oral administration with a safe dosage was only 0.043 times (0.017/0.395 lg/mL) lower than the concentration via IV injection with its LD50 dosage. However, the dosage of oral administration was more than about 1105 times (20.8/0.00904 g/kg) higher than the dosage of IV injection. This interesting phenomenon may demonstrate that the acute toxicity of a certain chemical compound not only correlates with its concentration in the blood after administration, but also relates to the type of administration route used to introduce the drug into the blood. This result also suggests that the absorption of BBR by the animal’s intestine system has its own limit, and no matter how much the orally administered dosage has increased, the absorption rate would not increase at this internal limit; the excess berberine will be excreted or removed by increasing metabolism (Zuo et al., 2006). In our results, we found that 20.8 g BBR/kg of body weight is safe for oral administration in mice. The mice also have a metabolic rate per kg body weight that is approximately (and conservatively) seven times higher than humans (Terpstra, 2001). Therefore, BBR enters the bloodstream much more quickly in mice. In light of this, the safety dosage for humans would be approximately 1/7 of 20.81 g BBR/kg (2.97 g BBR/kg of human body weight). We can reasonably infer from these data that the clinically recommended dosage of 15 mg BBR/kg human body weight (three 100 mg berberine tablets taken three times a day for the average 70 kg adult human) is safe. In the article, ‘‘Berberine is a novel cholesterol-lowering drug working through a unique mechanism distinct from statins” (Kong et al., 2004), we can see that the dosage used in the experiment by Kong and colleagues was safe. In the article, it states that the human participants were administered 0.5 g of BBR twice daily. Assuming the average human participant weighed on approximately 70 kg and that the 500 mg of BBR was given every 12 h, it is clearly well below the safety dosage of BBR. 5. Conclusion This is the first report that involves a thorough and well-conducted study of berberine’s acute toxicity via different injection
1109
routes and their corresponding blood concentrations systematically. The LD50 of BBR for IV and IP injections was determined for the first time in mice. Attempting to find the LD50 of BBR for IG administration proved to be strenuous and the LD50 could not be established, but the limitation of BBR blood concentration and its effect on mice mortality rate was explored. This study will be of great benefit to the understanding and practice of BBR use in vivo and in vitro.
Conflict of interest The authors declare that there are no conflicts of interest. Acknowledgment This project (90713043, 30801523, and 30973896) was supported by the National Natural Science Foundation of China. References Anis, K.V., Kuttan, G., Kuttan, R., 1999. Role of berberine as an adjuvant response modifier during tumour therapy in mice. Pharmaceutical Pharmacology Communications 5, 697–700. Chanter, D.O., Heywood, R., 1982. The LD50 test: some considerations of precision. Toxicology Letters 10 (2–3), 303–307. Chen, Y.Y., Wang, X.L., Sun, H., Xing, D.M., Hu, J., Wai, Z.H., Du, L.J., 2007. Characterization of the transportation of berberine in Coptidis rhizoma extract through rat primary cultured cortical neurons. Biomedical Chromatograph 22, 28–33. Hua, W.Y., Ding, L., Chen, Y., Gong, B., He, J.C., Xu, G.L., 2007. Determination of berberine in human plasma by liquid chromatography–electrospray ionization– mass spectrometry. Journal of Pharmaceutical and Biomedical Analysis 44, 931– 937. Ko, W.H., Yao, X.Q., Lau, C.W., Law, W.I., Chen, Z.Y., Kwok, W., Ho, K., Huang, Y., 2000. Vasorelaxant and antiproliferative effects of berberine. European Journal of Pharmacology 399, 187–196. Kong, W.J., Wei, J., Abidi, P., Lin, M., Inaba, S., Li, C., Wang, Y.L., Wang, Z.H., Si, S.Y., Pan, H.N., Wang, S.K., Wu, J.D., Wang, Y., Li, Z.R., Liu, J.W., Jiang, J.D., 2004. Berberine is a novel cholesterol-lowering drug working through a unique mechanism distinct from statins. Nature Medicine 10 (12), 1344–1351. Kong, W.J., Wei, J., Zuo, Z.Y., Wang, Y.M., Song, D.Q., You, X.F., Zhao, L.X., Pan, H.N., Jiang, J.D., 2008. Combination of simvastatin with berberine improves the lipid-lowering efficacy. Metabolism Clinical and Experimental 57, 1029– 1037. Kulkarni, S.K., Dhir, A., 2008. On the mechanism of antidepressant-like action of berberine chloride. European Journal of Pharmacology 589, 163–172. Kuo, C.L., Chou, C.C., Benjamin, Y.M., 1995. Berberine complexes with DNA in the berberine-induced apoptosis in human leukemic HL-60 cells. Cancer Letters 93, 193–200. Liu, J., He, C., Zhou, K., Wang, J., Kang, J.X., 2009. Coptis extracts enhance the anticancer effect of estrogen receptor antagonists on human breast cancer cells. Biochemical and Biophysical Research Communications 378 (2), 174– 178. Marinova, E.K., Nikolova, D.B., Popova, D.N., Gallacher, G.B., Ivanovska, N.D., 2000. Suppression of experimental autoimmune tubulointerstitial nephritis in BALB/C mice by berberine. Immunopharmacology 48, 9–16. Mirska, I., Kedzia, H., Kowalewski, Z., Kedzia, W., 1972. The effect of berberine sulfate on healthy mice infected with Candida albicans. Archive of Immunology Therapeutics Experiment 20, 921–929. Terpstra, A.H.M., 2001. Differences between humans and mice in efficacy of the body fat lowering effect of conjugated linoleic acid: role of metabolic rate. Journal of Nutrition 131, 2067–2068. Wang, Y.X., Yao, X.J., Tan, Y.H., 1994. Effects of berberine on delayed after depolarizations in ventricular muscles in vitro and in vivo. Journal of Cardiovascular Pharmacology 23, 716–722. Wang, X.L., Wang, R.F., Xing, D.M., Su, H., Ma, C., Ding, Y., Du, L.J., 2005. Kinetic difference of berberine between hippocampus and plasma in rat after intravenous administration of Coptidis rhizoma extract. Life Sciences 77, 3058–3067. Wang, X.Q., Qiu, S.J., Yao, X., Tang, T.T., Dai, K.R., Zhu, Z.A., 2009. Berberine inhibits staphylococcus epidermidis adhesion and biofilm formation on the surface of titanium alloy. Journal of Orthopaedic Research 27, 1487– 1492. Yin, J., Xing, H.L., Ye, J.P., 2008. Efficacy of berberine in patients with type 2 diabetes mellitus. Metabolism Clinical and Experimental 57, 712–717. Zhao, W., Zhou, Z.X., Kong, W.J., Jiang, J.D., 2008. Reduction of blood lipid by berberine in hyperlipidemic patients with chronic hepatitis or liver cirrhosis. Biomedicine and Pharmacotherapy 62, 730–731.
1110
M.M. Kheir et al. / Food and Chemical Toxicology 48 (2010) 1105–1110
Zhou, H., Mineshita, S., 2000. The effect of berberine chloride on experimental colitis in rats in vivo and in vitro. Journal of Pharmacology and Experimental Therapeutics 294, 822–829. Zhou, X.Q., Zeng, X.N., Kong, H., Sun, X.L., 2008. Neuroprotective effects of berberine on stroke models in vitro and in vivo. Neuroscience Letters 447, 31–36.
Zuo, F., Nakamura, N., Akao, T., Hattori, M., 2006. Pharmacokinetics of berberine and its main metabolites in conventional and pseudo germ-free rats determined by liquid chromatography/ion trap mass spectrometry. Drug Metabolism and Disposition 34 (12), 2064–2072.