An Improved HPLC-DAD Method for Quantitative Comparisons of Triterpenes in Ganoderma lucidum and Its Five Related Species Originating from Vietnam

An HPLC-DAD method for the quality control of wild and cultivated Ganoderma lucidum (Linhzhi) and related species samples was developed and validated. The quantitative determination of G. lucidum and its related species using 14 triterpene constituents, including nine ganoderma acids (compounds 4–12), four alcohols (compounds 13–16), and one sterol (ergosterol, 17) were reported. The standard curves were linear over the concentration range of 7.5–180 µg/mL. The LOD and LOQ values for the analyses varied from 0.34 to 1.41 µg/mL and from 1.01 to 4.23 µg/mL, respectively. The percentage recovery of each reference compound was found to be from 97.09% to 100.79%, and the RSD (%) was less than 2.35%. The precision and accuracy ranged from 0.81%–3.20% and 95.38%–102.19% for intra-day, and from 0.43%–3.67% and 96.63%–103.09% for inter-day, respectively. The study disclosed in detail significant differences between the quantities of analyzed compounds in different samples. The total triterpenes in wild Linhzhi samples were significantly higher than in cultivated ones. The total constituent contents of the five related Linhzhi samples were considerably lower than that in the G. lucidum specimens, except for G. australe as its constituent content outweighed wild Linhzhi’s content by 4:1.

of the G. lucidum species collected in Vietnam have not been reported, and no data exist that compare the components of wild-harvested, cultivated, and other related Linhzhi species from Vietnam. In this study, a reverse-phase HPLC method was developed for the fingerprint analysis and simultaneous determination of 17 compounds including a new lanostane triterpene (butyl lucidenate E2 (11) [19], uracil (1), 5-dihydrobenzoic acid (3, gentisic acid) [20], 12 lanostane triterpene derivatives (compounds 4-10 and 12-16), adenosine (2), and ergosterol (17). The developed method was successfully applied to the quantification of 14 triterpenoids in six wild and four cultivated G. lucidum samples, and five related Ganoderma species.

Optimization of Sample Preparation Condition
Several methods for the extraction of the fruiting bodies of Ganoderma species were surveyed including ultrasonication, refluxing, and maceration using methanol, but the ultrasonication method was the most effective, therefore we used this method to evaluate the effect of different solvents (100% methanol and 100% ethanol) on the amount of sample extracted. When 100% methanol was used, the content of the sample extracted was higher. To test the necessary time to accomplish the extraction, samples were prepared for 30, 60, 90, and 120 min. Since, the amount of the sample extracted after 90 min was same as the 120 min sample and higher than the 30 min sample, 90 min was selected as the optimal extraction time.

Selection of HPLC Conditions and Validation of the Developed Method
Although several HPLC methods have been reported for the determination of the constituents of Linhzhi samples [11,12,14,21,22], few constituents have been analyzed within the same study due to a lack of standard reference compounds. Our previous chemical investigation of G. lucidum from Vietnam resulted in the extraction and isolation of 17 compounds 1-17 ( Figure 1). The chemical structures of the isolated compounds were identified using UV-Vis, IR, 1 H-and 13 C-NMR, and mass spectrometry as well as by the comparison of these spectroscopic data with those reported in literature as a new lanostane triterpene (butyl lucidenate E2 (11) [19], adenosine (2), and two compounds isolated from the first time in G. lucidum, namely uracil (1) and gentisic acid (3) [20]. In addition, 11 lanostane steroids were identified, including lucidenic acid N (4), lucidenic acid E2 (5), ganoderic acid A (6), lucidenic acid A (7), ganoderic acid E (8), methyl lucidenate E2 (9), methyl lucidenate A (10), and butyl ganoderate A (12) [23], lucidadiol (13), ganodermanontriol (14), ganoderiol F (15), ganodermadiol (16), and ergosterol (17) [19,20,24]. The purities of the compounds were greater than 95%, as estimated using an HPLC-DAD method. Most Linhzhi triterpenoids contain a conjugated skeleton, and their UV absorption peaks are concentrated at 210, 237, 243, 253, and 255 nm [11,16]. The analytes were divided into three groups including lanostane triterpenoid-type alcohols and acids, sterol, and others including gentisic acid and adenosine. Based on the maximum absorption of the compounds, the detection wavelengths were set at 256 nm for the acids and their derivatives and 243 nm for the others. The retention times of the compounds in the analyzed samples were distinguished by comparing with those of each reference compound, which are shown in Table 1. The sample preparation conditions for the extraction of the compounds in the Ganoderma species were optimized, which are described in Section 2.1. This study describes the results of the fingerprint analysis of the compounds 1-17 and the quality analysis of 14 of them (compounds 4-17). The chromatographic fingerprints of the Ganoderma species are shown in Figure 2, which are divided into three groups including (A) the wild Linhzhi group, (B) the cultivated Linhzhi group, and (C) the related species group. This HPLC-DAD method was validated for linearity, the limits of detection (LOD) and limits of quantitation (LOQ), recovery, and reproducibility. Each coefficient of correlation (r 2 ) was >0.999, as determined by least square analysis, suggesting good linearity between the peak area ratio versus the compound concentration (Table 1). The LOD and LOQ were examined based on the lowest detectable peak in the chromatogram with a signal-to-noise (S/N) ratio of 3 and 10, respectively. Under our experimental conditions, we determined the LOD and LOQ for the 14 reference compounds in Table 2. The values obtained for both the LOD and LOQ in these analyses were low enough to detect traces of the compounds in the crude extract.   For the recovery, each reference compound was spiked into 1 g of each Ganoderma species at three levels, as described in the Experimental Section. The spiked samples were assayed, and the recoveries of each reference compounds were found to be 97.09% to 100.79%, and the relative standard deviation (RSD) (%) was less than 2.31% ( Table 2). The average recovery was represented by the formula: R (%) = [(amount from the sample spiked standard − amount from the sample)/amount from the spiked standard] × 100. Table 3 shows the intra-day and inter-day precision (%RSD) of this HPLC method. The precision and accuracy ranged from 0.81%-3.20% and 95.38%-102.19% for intra-day and from 0.40%-3.67% and 95.63%-103.09% for inter-day, respectively. The data demonstrate that the method was acceptable in terms of linearity, accuracy, and reproducibility.

Quantitative Comparison of Different Ganoderma Species
The amount of the chemical compounds within the samples was influenced by various factors such as the place of origin, type of study sample (cultivated or wild samples; different species of the same genus), and harvesting season. The variation of the lanostane triterpenoid alcohol or acid derivatives and ergosterol in the different Ganoderma species originating from the wild and cultivated collections from Vietnam was evaluated. The fingerprint analysis of the wild Linhzhi group ( Figure 2A) showed a similarity across the chromatograms, and the 17 analytes were present in all the samples. Figure 2B,C show the differences among the wild Linhzhi group (A group), the cultivated Linhzhi group (B group), and the related Linhzhi species group (C group). In particular, the chromatograms of the five related Linhzhi species including G. sp, G. applanatum, G. australe, G. colossum, and G. subresinosum showed obvious differences. Compounds 1-3 were not distinctly separated using the developed method, so they were not quantified. This study focused on the simultaneous determination of the remaining 14 compounds 4-17 by using the developed HPLC-DAD method for the all samples (Groups A, B and C), which are summarized in Tables 4-6, respectively. Each sample was analyzed in triplicate to ensure the reproducibility of the quantitative results. The comparison of the 14 compounds in the Linhzhi samples (both wild and cultivated samples) showed that the number of acids and their derivatives in all the samples is significantly higher than the number of alcohols. However, there are differences between the wild and cultivated samples. While the total amount of the acids and alcohols in the wild samples vary from 2089.40 to 44,703.07 μg/g and 917.41 to 2498.68 μg/g, respectively, those from the cultivated samples fluctuate between 1003.83 and 1720.69 μg/g and 153.31 to 549.32 μg/g, respectively. Similarly, the total amount of the acids in G. australe outweighs that of the wild and cultivated Linhzhi samples and other related Linhzhi species (Tables 4-6,) with a total amount of 19,999.28 μg/g. In addition, the amount of the 2 compounds 11 and 12 in the analyzed samples was below LOQ except for TGau, which had 131.29 ± 1.31 µg/g.      Tables 4-6 and Figure 3, the total amount of all the compounds in the wild samples was appreciably higher than in the cultivated ones. For example, the amount of compound 4 (lucidenic acid N) in the wild G. lucidum sample varied from 257.80-845.46 μg/g; however, this compound was not observed in GL2, and in the other cultivated samples (Group C) it fluctuated between 52.53-139.08 μg/g. Another good example is lucidenic acid E2 (5), which was found in the wild G. lucidum samples in a range from 319.47 to 1,766.75 μg/g in comparison with the cultivated G. lucidum samples in a range from 258.06 to 481.31 μg/g. In addition, the wild samples contained significantly more ganoderiol F (15) than the cultivated samples, which varied between 563.94 μg/g and 1,635.06 μg/g and between 65.03 μg/g and 226.71 μg/g, respectively. Interestingly, a different trend was observed for methyl lucidenate E2 (9), which was under the LOQ in the wild samples but was found in GL1, GL2, and GL3 in the range between 286.94 and 446.95 μg/g. On the whole, the samples from Bac Giang (VN16 and VN18) seemed to have a higher amount of the constituents than the specimens from Quang Nam (VN1, VN12, and VN13). The total amount of constituents in the related Linhzhi samples (Group C) including in TGau, TGLs, TGap, TGc, and TGs was considerably lower than the amount in the Linhzhi samples of Groups A and B. The amount of constituents in TGau outweighed that of the others, as it was about 4 times as high as that of the wild Linhzhi samples (VN12). The proportion of constituents in the other species is different from G. lucidum. More specifically, the proportion of lucidenic acid E2 (5), which is one of the major compounds in G. lucidum, is low in TGap and TGS and is not found in TGau. Similarly, while almost G. lucidum samples contain a considerable amount of ganodemanontriol, it was only seen in trace amounts in the TGLs, TGap, TGau and TGs samples. In contrast, the amount of methyl lucidenate E2 (9), which is not observed in G. lucidum, is the major compound in the TGap, TGau, and TGc samples. It is noteworthy that the amount of the constituents in TGs was substantially smaller than others, and 50 percent was ergosterol.

Discussion
To date, several previous studies have reported using HPLC analytical methods for the analysis of Ganoderma lucidum and its related products. For example, Zhao et al. used HPLC for the determination of 9 triterpenes and sterols for the quality evaluation of G. lucidum [25]. In a study from Wang et al., an RP-HPLC method was developed for the determination of six ganoderic acids [22]. In 2004, Gao and coworkers reported the quantitative determination of 19 triterpene constituents, including six ganoderma alcohols and 13 ganoderic acids [11]. These studies and others focused only on the ganoderic acids and their derivatives [12,26]. Nucleosides, nucleobases, and polysaccharides were used for the qualitative and quantitative analyses of Ganoderma spp [27]. However, these studies are insufficient for a comprehensive evaluation of G. lucidum, and there is little data that compare G. lucidum from different origins or compare G. lucidum and its related species.
In this paper, we developed and optimized an HPLC-DAD method that allows for the specific identification of many terpenes. Fourteen triterpenes, including nine ganodermic acids 4-12, four alcohols 13-16, and one sterol (ergosterol, 17), were used for the quantitative determination of G. lucidum and its related species. Eight of the nine ganoderma acids (all but ganoderic acid A) had never been analyzed before. Two new lanostane triterpenes 11 and 12, which recently were discovered by our group and Lee Iksoo et al. [24], were used in the fingerprint analysis and quantitative determination for the first time [19,23]. Alcohols 13 and 16, which had never been examined quantitatively in previous studies, were used to evaluate G. lucidum and its related species chemically using HPLC-DAD. Two Ganoderma species investigated in this study were studied quantitatively using HPLC for the first time, except for G. applanatum [28]. However, in a study by Liu et al., G. applanatum was evaluated by using five ganoderic acids. Moreover, two compounds, uracil and gentisic acid, which were found in G. lucidum for the first time, were confirmed using the HPLC fingerprint technique.
In comparison with the results of previous studies, this study showed both similarities and differences. In a study by Gao et al. [11], the amount of ganondermanontriol (14) in Japanese Linhzhi ranged widely from 19.2 to 235.3 μg/g. In our study, there was a wide variation of compound 14 in the wild, cultivated, and related Linhzhi samples ranging from 129.31 to 394.10 μg/g, 50.85 to 208.34 μg/g, and 72.99 ± 1.88 μg/g (related Linhzhi species, TGc), respectively. Therefore, the amount of compound 14 in the Japanese samples was similar to the Vietnamese cultivated samples but was lower than that found in the Vietnamese wild Linhzhi. These results indicate that the amount of compound 14 may not depend on geographic factors but instead is affected by the cultivation conditions. Gao's study showed that the contents of ganoderiol F (15) ranged from 18.9 to 156.5 μg/g [11]. However, the Vietnamese wild samples contained 563.94-1635.06 μg/g of 15, and the Vietnamese cultivated samples contained 65.03-226.71 μg/g of 15. Both the wild and cultivated Linhzhi samples from Vietnam contain more compound 15 than the Linhzhi from Japan. Similar to the results found with compound 14, these results indicate that amount of 15 is not only affected by geographic factors but also by cultivation conditions. In a study from the Yuan group, the content of ergosterol (17) from sporoderm-broken germinating spores of Linhzhi varied from 32 μg/g to 1202 μg/g in the cultivated Linhzhi from China [21], corresponding with the results from this study, as the content of 17 in the cultivated Linhzhi ranged between 135.14 μg/g to 795.96 μg/g. With regard to methyl lucidenate E2 (9), there was a huge difference in the amounts found among the wild and cultivated Linhzhi and its related species. While a quantitative determination of 9 in the wild species could not be made, its content was above 288 μg/g in the cultivated species. Interestingly, compound 9 was the major compound in related species of Linhzhi as its content was at least 1023.84 μg/g across the species and was as high as 2499.52 μg/g in G. australe.

Chemicals and Reagents
Acetonitrile and methanol (MeOH) of analytical HPLC grade was purchased from Merck (Darmstadt, Germany). Phosphoric acid of analytical reagent grade was obtained from Sigma-Aldrich (St Louis, MO, USA). The other organic solvents and other chemical reagents were of analytical reagent grade.

Reference Compound Preparation
To determine the content of fourteen markers (compounds 4-17) of Linhzhi and related Linhzhi samples, the dried powders were used for extraction. The same amounts (about 1 g) of pulverized fruiting bodies were weighed and sieved through 50 mesh and then placed into a volumetric flask, methanol (10 mL) was added, the weight was accurately measured and the samples were ultrasonically extracted for 90 min at 50 °C. The solution was cooled, weighed again, and made up the loss in weight with methanol. The solution was filtered through 0.45 µm membrane filter prior to HPLC analysis.

HPLC
Analytical HPLC was carried out on a LC 20A system (Shimadzu, city, Japan) consisting of a LC-20AD quaternary gradient pump, an autosampler, and a SPD-M20A diode array detector, connected to a LC solution singer ver. 1.25 software. A Zorbax XDB C18 (4.6 × 250 mm, 5 µm, Agilent Technologies, Inc., Santa Clara, California, CA, USA) was used. A binary gradient solution system consisted of 0.1% phosphoric acid in water (A) and acetonitrile (B) and separation was achieved using the following gradient program: 0 min, 4% B; 10 min, 11% B; 15 min, 30% B; 60 min, 45% B; 90 min, 85% B; 110 min, 100 B%; 130-140 min, 100% B; and finally, reconditioning the column with 4% B isocratic for 10 min. The flow rate was 0.5 mL/min, the system operated at 40 °C and the detection wavelengths were set at 243 and 256 nm for ganoderma alcohols and acids, respectively.

Method Validation
Every standard compound was accurately weighed and dissolved in 100% MeOH to prepare a stock solution of 1.0 mg/mL concentration. Working standard solutions of ganoderma alcohols and acids were prepared by repeated dilution to give eight respective concentrations with methanol (7.5-180 µg/mL). Eight concentrations of 14 analyses were injected in triplicate, and then the calibration curves were constructed by plotting the peak areas versus the concentrations of each analysis. The linearity was demonstrated by a correlation coefficient (r2) greater than 0.999. The limit of detection (LOD) and the limit of quantification (LOQ) were determined based on signal-to noise ratios (S/N) of 3:1 and 10:1, respectively. Intra-and inter-day variations were chosen to determine the precisions of the developed method. The relative standard deviation (RSD) was taken as a measure of precision. Intraand inter-day repeatability was determined on five times within one day and five separate days, respectively. The recovery tests were prepared by mixing a powdered sample (1 g) with three concentration levels (25%, 50%, and 100%) of each compound. The mixture was then extracted by following the section of preparation of sample solution for HPLC analysis. The extract solutions were filtered through a 0.45 µM membrane. The HPLC-DAD analysis experiments were performed in triplicate for each control level. Precision were determined by multiple analysis (n = 5) of quality control samples. All samples were then subjected to HPLC analysis to calculate the recovery rates.

Statistical Analysis
The data were analyzed using the unpaired Student's t-test between the control and compounds. Data compiled from three independent experiments and values are expressed as mean ±SD.

Conclusions
This is the first time a HPLC-DAD method for quantitative analysis of constituents in Ganoderma lucidum and its related species batches originating from Vietnam was established. In the present work, we have reported for the first time the presence of lanostane triterpenes, ergosterol, uracil, adenosine, and gentisic acids in the Vietnamese G. lucidum and its related species. Especially, two new lanostanes, butyl lucidenate E2 and butyl ganoderate A, were reported for the first time in G. lucidum originating from Vietnam and its four related species using a HPLC-DAD method. In addition, the highest content of methyl lucidenate E2 was found in G. australe, G. applatatum, and G. colossum, respectively. In the present study the profile of the 17 compounds differed significantly in the all analyzed samples. It can be also concluded that the geographical distributions, growth conditions, and substrates might be the key to differences in producing chemical compositions. This present work suggested an accurate and sufficient method for quantitative evaluation, which is suitable for quality evaluation of Ganoderma products.