Rapid determination of lovastatin in the fermentation broth of Aspergillus terreus using dual-wavelength UV spectrophotometry

Abstract

Context: Lovastatin, a hypocholesterolemic drug, is produced by submerged fermentation of Aspergillus terreus Thom (Trichocomaceae). High performance liquid chromatography is usually used to determine lovastatin in samples of the fermentation broth. However, this method is inconvenient and costly, especially in the context of high-throughput sample analysis.

Objective: A direct and simple dual-wavelength ultraviolet spectrophotometric method for quantifying lovastatin in the fermentation broth of A. terreus was developed.

Materials and methods: A. terreus Z15-7 was used for all experiments. The liquid fermentation was conducted at 30 °C in a rotary shaker at 150 rpm for 15 d. Silica gel and neutral alumina column chromatography were used for the separation and purification of lovastatin from the fermentation broth.

Results: The limits of detection of lovastatin were 0.320 μg/ml in the lovastatin standard solution and 0.490 μg/ml in the fermentation broth sample and the limits of quantification of lovastatin were 1.265 μg/ml in the lovastatin standard solution and 3.955 μg/ml in the fermentation broth sample. The amounts of lovastatin in the fermentation broth ranged from 876.614 to 911.967 μg/ml, with relative standard deviations from 1.203 to 1.709%. The mean recoveries of lovastatin using silica gel and neutral alumina column chromatography were 84.2 ± 0.82 and 87.2 ± 0.21%, respectively.

Discussion and conclusion: Dual-wavelength UV spectrophotometry is a rapid, sensitive, accurate, and convenient method for quantifying lovastatin in fermentation broth. Neutral alumina column chromatography is more efficient than silica gel column chromatography for the purification and determination lovastatin using the developed dual-wavelength UV spectrophotometry method.

• Column chromatography
• hypocholesterolemic drug
• infrared spectrum
• neutral alumina column
• silica gel column
• ultraviolet spectrum

Introduction

Lovastatin is a fungal metabolite that is a competitive inhibitor of 3-hydroxyl-3-methylglutaryl coenzyme A (HMG-CoA) reductase, which is the rate-limiting enzyme in cholesterol biosynthesis (Alberts et al., 1980). Lovastatin can effectively reduce plasma cholesterol levels in various mammalian species, including humans, and is thereby effective in the treatment of hypercholesterolemia (Frishman et al., 1989). Lovastatin was the first hypocholesterolemic drug to be approved by the United States Food and Drug Administration. It is used preventatively in cardiovascular disease and reduces the risk of morbidity and death from coronary artery disease. Moreover, lovastatin is used as a precursor for the manufacture of simvastatin, which is a more potent semisynthetic derivative of lovastatin. In addition, it has been demonstrated that lovastatin has biological effects beyond reducing cholesterol levels, including stabilizing atheromatous plaques, modifying atherosclerosis progression, improving endothelial functions, modulating inflammatory responses, and preventing thrombus formation (Barrios-González & Miranda, 2010). Currently, lovastatin is primarily commercially produced via submerged fermentation of Aspergillus terreus.

Dual-wavelength spectrophotometry methods can be used to determine an unknown concentration of a component of interest that is present in a mixture containing both the component of interest and an unwanted interfering component by determining the difference in absorbance between two points in the spectrum of the mixture (Chance, 1972). This method has been useful for the measurement of small changes in absorbance in systems containing high-absorbing backgrounds (Cowles, 1965). It was also a promising analytical technique for suspension analysis (Merrick & Anthony, 1989). The prerequisite for the application of a dual-wavelength method is the selection of two wavelengths such that the interfering component shows the same absorbance while the component of interest shows a significant difference in absorbance with changes in concentration (Chance, 1972). The technique consists of using a reference wavelength to correct for a highly absorbing background. In addition, dual-wavelength spectrophotometry requires extremely small quantities of sample for analysis.

A number of analytical techniques have been applied for the determination of lovastatin in aqueous and biological samples, including gas chromatography-mass spectrometry (GC-MS; Morris et al., 1993; Wang-Iverson et al., 1989), high performance liquid chromatography (HPLC; Zhang & Yang, 2007) and reversed-phase HPLC (Ye et al., 2000) with ultraviolet detection (LC-UV), liquid chromatography-mass spectrometry (LC-MS) and LC-MS-MS (Miao & Metcalfe, 2003), spectrofluorimetry (Mabrouk et al., 1998), and liquid chromatography-electrospray ionization–mass spectrometry (LC-ESI-MS; Li et al., 2008). HPLC has been the most commonly used method for the determination of lovastatin in fermentation broth (Kysilka & Kren, 1993; Sorrentino et al., 2010). Although HPLC is a highly sensitive method, it requires samples pretreatment for the determination of lovastatin in fermentation broth, which is inconvenient and time-consuming, especially for screening mutant strains and optimizing culture conditions. Therefore, it is necessary to develop a rapid, accurate, and convenient method for the determination of lovastatin. This investigation was initiated to highlight the applicability of dual-wavelength spectroscopy and to use the developed method to quantify the concentration of lovastatin in the fermentation broth of A. terreus. The objective of this work was to develop a direct and simple dual-wavelength ultraviolet spectrophotometric method for the determination of lovastatin in fermentation broth. The performances of silica gel and neutral alumina column chromatography in the separation and purification of lovastatin from fermentation broth were also compared.

Materials and methods

Chemicals, apparatus, and conditions

A lovastatin standard (internal standard, purity ≥99%) was acquired from the National Institute for the Control of Pharmaceutical and Biological Products (Beijing, China). Neutral aluminum oxide (for TLC) and silica gel (ZCX-II) were obtained from commercial sources and used within their shelf life period. All reagents and chemicals used were of analytical grade (Shanghai Chemical Reagent Corporation, Shanghai, China). Distilled water was doubly distilled in the laboratory and used throughout the study. Spectrophotometric analyses were performed with a microcomputer-based UV759S UV-visible double beam spectrophotometer (Shanghai Precision & Scientific Instrument Co Ltd, China) with a pair of 1 cm matched quartz cells. The settings used were as follows: 2 nm slit width, 0.5 nm wavelength accuracy, 5 s response time, and a scan speed of 60 nm/min. Chromatograms were monitored at 238 nm and the temperature of the column was maintained at room temperature.

Standard solution preparation

The lovastatin stock solution was prepared by dissolving an accurately weighed amount (3.2 mg) of lovastatin standard in 50 ml of 75% ethanol and stored in a brown glass bottle. This stock solution was stored at 4 °C until used to prepare working solutions by adding an appropriate volume of 75% ethanol. Working solutions of different concentrations were prepared from the stock solution immediately prior to use. For the construction of a calibration curve, a 64 μg/ml of lovastatin standard solution was prepared. First, 3.2 mg of the lovastatin standard was accurately weighed into a 50-ml volumetric flask and dissolved in a diluents consisting of 75% ethanol. Then, 0.5, 1.0, 2.0, 3.0, 4.0, 5.0, 6.0, 7.0, 8.0, and 9.0 ml of the standard solutions were diluted to a constant volume of 10 ml, and the final concentrations were 3.2, 6.4, 12.8, 19.2, 25.6, 32.0, 38.4, 44.8, 51.2, and 57.6 μg/ml, respectively.

Strain, culture medium, and fermentation conditions

The strain used in the experiments was A. terreus Z15-7, which is a high-producing lovastatin mutant identified in our previous study (Li et al., 2011). Seed cultures were prepared by transferring an inoculating shovel of spores from a malt extract agar slant into a 100-ml conical flask containing 50 ml of basal medium (50 g glucose, 10 g corn meal, 5 g peptone, 2 g NaCl, 0.5 g MgSO₄·7H₂O, and 0.5 g KH₂PO₄ in 1000 ml of distilled water, adjusted to pH 6.0), followed by incubation at 30 °C in a rotary shaker at 150 rpm for 24 h. Then, fermentations were performed by inoculating 5 ml of the seed broth into 50 ml of fermentation medium consisting of 50 g glycerol, 10 g corn meal, 2 g NaCl, 2 g NaNO₃, 0.5 g MgSO₄·7H₂O, and 0.5 g KH₂PO₄ in 1 000 ml of distilled water (adjusted to pH 6.0) followed by incubation at 30 °C in a rotary shaker at 150 rpm for 15 d (Li et al., 2011).

Sample pre-treatment

After 15 days of incubation, the culture broth was harvested and adjusted to pH 3.0 using HCl. Next, an equal volume of ethyl acetate was added. After shaking at 180 rpm for 12 h at ambient temperature, the fermentation broth and the mycelium pellets were filtered through membrane filters and the residual mass was washed three times with ethyl acetate. The organic and aqueous phases from the filtrate were separated using a separating funnel. The organic phase was dried in a rotary evaporator under vacuum at 45 °C. The dried residue was dissolved in 5 ml of 75% ethanol and used for the determination and purification of lovastatin.

Chromatography

Silica gel and neutral alumina column chromatography, used for the separation and purification of lovastatin from the fermentation broth, were performed at ambient temperature. For silica gel column chromatography, the dried lovastatin residue was dissolved in a small amount of acetone and mixed with 1 g of silica gel that was pretreated and soaked with petroleum ether. The mixture was packed into a chromatography column and eluted with a mixture of petroleum ether and acetone (7:1, v/v). For neutral alumina column chromatography, the dried lovastatin residue was dissolved in 2 ml of acetone and mixed with 1 g of neutral alumina that had been pretreated at 100 °C for 30 min. The mixture was added to a chromatography column and eluted with 95% ethanol. Meanwhile, a lovastatin standard or lovastatin standard mixed with a sample from the fermentation broth were added separately into two kinds of columns. The eluents were collected using an automatic fraction collector. After detection using thin-layer chromatography, the eluents containing lovastatin were dried in a rotary evaporator under vacuum at 45 °C. The dried residue was dissolved in 5 ml of 75% ethanol and the lovastatin content was determined using dual-wavelength UV spectrophotometry.

Dual-wavelength UV spectrophotometry and infrared spectroscopy

Lovastatin contents were determined using the dual-wavelength UV spectrophotometry method. The spectrum of lovastatin was recorded in absorbance mode. The radiation source was a deuterium lamp emitting a continuous UV spectrum between 226 and 260 nm. The beginning and ending wavelengths were 226 and 260 nm, respectively. The absorbencies at 246 nm, which is the characteristic absorption peak of lovastatin, and at 254 nm were recorded and used to plot a calibration curve for calculating the lovastatin content. The instrument noise of the UV759S spectrometer was 0.0078 against 75% ethanol.

The infrared (IR) spectra of the sample and the lovastatin standard were recorded in potassium bromide disks at ambient temperature using a Vertex 70/70v FT-IR spectrometer (Bruker Co., Germany).

Validation of the method

The linearity of the proposed method was verified by analyzing ten different concentrations in the range of 3.20 to 64.0 μg/ml. Each concentration was analyzed against the blank (75% ethanol) in triplicate. The difference in absorbance between 246 and 254 nm was calculated for each sample, and the results were used to plot a calibration curve. Equations were estimated from the calibration curves using linear regression analysis. Correlation coefficients were calculated and statistical evaluation of the described method was performed.

The precision of the method was determined by performing intra- and inter-day variation and method repeatability studies. The intra-day precision of the method was evaluated by analyzing samples of three different concentrations of lovastatin (9.04, 18.08, and 36.16 μg/ml) in triplicate on the same day. The inter-day precision was evaluated using samples of the same concentrations on three consecutive days. Precision was evaluated using samples of the same concentrations that were analyzed by three different analysts, and the relative standard deviation (RSD) was calculated.

To validate the accuracy of the proposed method, the three nominal concentrations of lovastatin (9.04, 18.08, and 36.16 μg/ml) were assayed in triplicate. The accuracy (% bias) was calculated from the nominal concentration and the mean value of the measured concentration.

The limit of detection (LOD) was determined as the lowest concentration of lovastatin in a standard or fermentation broth sample that had distinguishable absorption peaks in the UV spectrum. The limit of quantification (LOQ) was determined as the lowest concentration of lovastatin in a standard or fermentation broth sample that could be quantified with acceptable precision and accuracy under the experimental conditions with a signal-to-noise ratio of at least 10:1.

To determine the recovery of the method, a known amount of lovastatin was mixed with the unknown pigment separated from the fermentation broth using column chromatography and then the absorbencies were measured and the contents of lovastatin were calculated.

Results and discussion

Dual-wavelength UV spectrophotometry-based determination

The results of scanning the UV spectra of lovastatin standard, pigment, and ethyl acetate extract extracted from the fermentation broth (in 75% ethanol) are shown in Figure 1. The lovastatin standard had three characteristic absorption peaks at 230, 238, and 246 nm, which was consistent with an earlier report from Endo (1979). We also acquired UV spectra of the ethyl acetate extracts from the fermentation broth as well as the spectrum of an unknown pigment that was separated from the ethyl acetate extract using silica gel chromatography. The results indicated that the ethyl acetate extract has the same UV absorption peaks as the lovastatin standard. The unknown pigment also shows strong absorption from 226 nm to 260 nm. In theory, the highest absorption peak of lovastatin at 238 nm should be the most sensitive wavelength to use in spectrophotometric assays. In practice, none of the absorbance at 238, 230, or 246 nm can be directly used to determine the lovastatin in fermentation broth because the unknown pigment that is also present has strong absorbency at these wavelengths. We determined that the absorbance of the lovastatin standard at 254 nm is nearly equal to 0; therefore, the difference in the values between the absorbance at 254 and 246 nm is equal to the absorbance at 246 nm, and it is just one of the characteristic absorption peaks of lovastatin. Furthermore, the difference between the absorbance of the pigment at 254 nm and at 246 nm is near to zero (i.e., A₂₄₆ − A₂₅₄ ≈ 0). According to the principles of dual-wavelength spectrophotometry, the wavelengths of 246 nm and 254 nm were used in further experiments. The absorbances of lovastatin at 246 and 254 nm were monitored separately, and the A₂₄₆ − A₂₅₄ value was used to calculate the contents of lovastatin in samples.

Figure 1. UV spectra of the lovastatin standard, pigment, and ethyl acetate extract from the fermentation broth (in 75% ethanol).

Linearity, precision, accuracy, recovery, and sensitivity of the method

The calibration curve of the analytical method was constructed by plotting concentration versus absorbance, and it showed suitable linearity in the range 3.2–57.6 μg/ml. The representative linear equation was Y = 0.0276X + 0.0054, with a highly significant correlation coefficient (r = 0.9994). The calibration curve was validated by ANOVA, which indicated significant linear regression.

The precision of the measurements was determined using the three standard samples of lovastatin (Table 1). The inter-assay RSDs of the three samples ranged from 0.405 to 2.390%, and the interassay RSDs ranged from 0.833 to 1.470%. The accuracy and recovery of the measurements were determined using the five standard samples of lovastatin (Table 2). The % bias of the five standard samples of lovastatin ranged from 0.075 to 1.013%. The average recoveries ranged from 100.1 to 100.97%, and the RSDs ranged from 0.137 to 2.155%. The theoretical limit of detection (LOD) was determined as the lowest concentration of lovastatin in the standard and in the standard mixed with the pigment separated from the fermentation broth that had a strong absorbency within the UV scanning spectra. The results demonstrated that a distinguishable absorption peak at 246 nm was detected when the concentrations of lovastatin were 0.320 and 0.490 μg/ml in the standard and in the standard mixed with the pigment, respectively (Figure 2). This result indicated that the limit of detection of lovastatin in fermentation broth should theoretically be 0.490 μg/ml. The results of the limit of quantification (LOQ) assay showed that the RSD and bias were 17.126 and 31.953%, respectively, when the concentration of the lovastatin in standard was 0.633 μg/ml. When the concentration of lovastatin was 1.978 μg/ml in the standard mixed with the pigment, the RSD and bias were 18.622 and 35.777%, respectively (Table 3). Furthermore, the RSD and bias were 4.857 and 7.431%, respectively, when the concentration of the lovastatin standard was 1.265 μg/ml. When the concentration of lovastatin was 3.955 μg/ml in the standard mixed with the pigment, the RSD and bias were 8.40 and 13.477%, respectively (Table 3). These results indicated that the limits of quantification of lovastatin were 1.265 and 3.955 μg/ml in the standard solutions and the fermentation broth, respectively.

Figure 2. The limit of detection of the lovastatin standard (A) and the nominal concentrations of lovastatin mixed with the pigment isolated from the fermentation broth (B).

Table 1. Precision of lovastatin determination.

	Nominal concentration (μg/ml)	Observed concentration (μg/ml)	Standard deviation	RSD%
Intra-assay	9.040	9.162 ± 0.161	0.219	2.390
	18.080	18.176 ± 0.022	0.132	0.726
	36.160	36.581 ± 0.091	0.148	0.405
Inter-assay	9.040	9.049 ± 0.067	0.133	1.470
	18.080	18.357 ± 0.075	0.153	0.833
	36.160	36.654 ± 0.023	0.452	1.233

Observed concentration data are expressed as mean ± SE (n = 3) (μg/ml).

Table 2. Accuracy and recovery of lovastatin determination.

Nominal concentration (μg/ml)	Observed concentration (μg/ml)	Standard deviation	Average recovery%	RSD%	bias%
9.040	9.069 ± 0.074	0.196	100.32	2.155	0.320
18.080	18.265 ± 0.056	0.168	100.97	0.920	1.013
36.160	36.454 ± 0.176	0.346	100.81	0.949	0.806
100	100.670 ± 0.080	0.138	100.67	0.137	0.667
200	200.150 ± 0.963	1.669	100.10	0.834	0.075

Observed concentration data are expressed as mean ± SE (n = 3) (μg/ml).

Table 3. Limit of detection and limit of quantification of lovastatin standard and lovastatin standard mixed with fermentation pigment.

	Nominal concentration (μg/ml)	Observed concentration (μg/ml)	Standard deviation	RSD%	bias%
Lovastatin standard	0.317	0.458 ± 0.009	0.100	21.834	44.479
	0.633	0.835 ± 0.008	0.143	17.126	31.953
	1.265	1.359 ± 0.017	0.066	4.857	7.431
	2.530	2.637 ± 0.046	0.076	2.883	4.209
	5.060	4.910 ± 0.071	0.106	2.159	−2.964
	10.120	9.799 ± 0.084	0.227	2.317	−3.172
	20.250	19.738 ± 0.096	0.362	1.834	−2.528
	40.500	40.755 ± 0.215	0.180	0.442	0.630
Lovastatin standard + pigment	0.494	1.231 ± 0.002	0.521	42.323	149.190
	0.989	1.748 ± 0.063	0.537	30.721	76.744
	1.978	2.685 ± 0.061	0.500	18.622	35.777
	3.955	4.488 ± 0.057	0.377	8.400	13.477
	7.910	7.991 ± 0.029	0.057	0.713	1.024
	15.820	15.982 ± 0.074	0.115	0.720	1.024
	31.640	30.810 ± 0.485	0.587	1.905	−2.623

Observed concentration data are expressed as mean ± SE (n = 3) (μg/ml).

Assay of lovastatin in fermentation broth

The developed method was used to determine lovastatin in the fermentation broth of A. terreus. Four batches of fermentation broth were assayed (n = 6). The lovastatin concentrations in the fermentation broths ranged from 876.614 to 911.967 μg/ml, and the RSDs ranged from 1.203 to 1.709%.

The dual-wavelength UV spectrophotometry method has been well studied (Merrick & Anthony, 1989). This method was used to detect amino acids in plasma by Palego (2010). Many precise analytical methods have been used for the determination of lovastatin in various biological samples, such as human plasma (Li et al., 2008; Mabrouk et al., 1998; Morris et al., 1993; Palego et al., 2010; Ye et al., 2000), human serum (Wang-Iverson et al., 1989), dosage forms (Mabrouk et al., 1998), drugs (Merrick & Anthony, 1989; Miao & Metcalfe, 2003), and rat tissue (Zhang & Yang, 2007). However, HPLC has been generally used for the quantification of lovastatin in fermentation broth (Friedrich et al., 1995; Kysilka & Kren, 1993; Sorrentino et al., 2010). Although, HPLC is a precise method for the determination of lovastatin, it is also a costly, inconvenient and time-consuming method. The present experiments demonstrated that UV spectrophotometry is a simple, inexpensive, reliable, and suitable method for the quantification of lovastatin in fermentation broth. We have successfully used this method to determine lovastatin concentrations in the fermentation broth during breeding of a high lovastatin-producing strain (Li et al., 2011).

Column chromatography of lovastatin

To evaluate the efficiency of silica gel and neutral alumina column chromatography for the separation and purification of lovastatin from fermentation broth, both columns were employed for purifying lovastatin in the lovastatin standard, the ethyl acetate extracts of the fermentation broth, and the lovastatin standard mixed with ethyl acetate extracts of the fermentation broth, respectively. The elution profiles are shown in Figures 3 and 4. The elution peaks of lovastatin in the three samples all appeared between 8 and 16 ml of eluent for the silica gel column, and lovastatin was detected in the eluent from 6 to 36 ml (Figure 3). The mean recovery was 84.2 ± 0.82% for all three samples. The elution peaks of lovastatin in the three samples appeared between 6 and 12 ml of eluent for the neutral alumina column, but lovastatin was barely detected in the eluent after 12 ml (Figure 4). The mean recovery was 87.2 ± 0.21% for all three samples. Moreover, more than three-times the volume of eluent was required to elute lovastatin from the silica gel column than from the neutral alumina column. Additionally, the petroleum ether and acetone (7:1, v/v) used for eluting the silica gel column had to be desiccated prior to quantitative measurement of lovastatin using the developed method. The characterized IR absorption peaks of lovastatin are at 3550, 2970, 1696, and 1220 cm⁻¹ in KBr (Endo, 1979). In the present study, the IR spectra of the lovastatin standard were 3541.2, 2964.5, 1699.2, and 1220.9 cm⁻¹. The IR spectra of lovastatin samples purified using alumina column were 3541.2, 2964.5, 1701.2, and 1220.9 cm⁻¹. The IR spectra of lovastatin samples purified using silica column were 3539.3, 2964.5, 1701.2, 1219.0 cm⁻¹ (Figure 5). These results indicated that lovastatin in the fermentation can be efficiently purified using neutral alumina column chromatography and silica column chromatography and the former is more efficient for the purification and quantification of lovastatin using the developed method.

Figure 3. Elution curve of lovastatin from silica gel column chromatography (the eluent was petroleum ether and acetone 7:1 v/v). A 30 cm × 1.5 cm column was used. The flow rate was 0.7 ml/min, and 2 ml fractions were collected.

Figure 4. Elution curve of lovastatin from neutral alumina column chromatography (the eluent was 95% ethanol). A 30 cm × 1.5 cm column was used. The flow rate was 0.7 ml/min, and 2 ml fractions were collected.

Figure 5. IR spectra (KBr) of the lovastatin standard and lovastatin isolated from the fermentation broth using column chromatography.

Lovastatin has been extracted from fermentation broth using methanol (Sorrentino et al., 2010; Valera et al., 2005) or ethyl acetate (Ali et al., 2007). It was necessary to centrifuge or filter the extract with a 0.2–0.45 μm membrane filter prior to HPLC analysis (Ali et al., 2007; Sorrentino et al., 2010; Valera et al., 2005). We utilized ethyl acetate to extract lovastatin from the fermentation broth of A. terreus and developed a column chromatography purification method (Li et al., 2011).

Conclusions

A convenient, rapid and sensitive dual-wavelength UV spectrophotometry method for the quantitative determination of lovastatin in the fermentation broth of A. terreus has been developed and validated. The method is simple and demonstrated good reproducibility and accuracy, and it was successfully utilized to assay the production of lovastatin in the fermentation broth. Additionally, this method is suitable for efficiently screening strains for lovastatin production. In comparison with previously developed HPLC methods, two major features of this technique should be highlighted: (1) it is very efficient and relies on a straightforward sample preparation procedure and (2) it offers comparable accuracy, a short analysis time with good resolution, sensitivity, great reproducibility, and requires a low-cost apparatus and conditions. This technique meets the requirements of high throughput strain screening for lovastatin production. Neutral alumina column chromatography has advantages over silica gel column chromatography in the purification and quantification of lovastatin from fermentation broth. The recoveries of lovastatin from lovastatin standards, the fermentation broth, and a mixture of fermentation broth with the lovastatin standard were 87.1, 87.6, and 86.9%, respectively.

Declaration of interest

The authors declare no conflicts of interest. The authors alone are responsible for the content and writing of this article. This work was supported by the National Natural Science Foundation of China (31260090 and 30960063).