Quality evaluation of commercial products of Ganoderma lucidum made from its fruiting body and spore (2024)

Materials

The standard extracts of GLS and GLF were obtained from the National Institute of Food and Drug Control (Beijing, China), which is used as the standard products for constructing the standard chromatogram for quality control.

The commercial products were provided by Ton Ren Tang (hereinafter referred to as TRT products), including nine batches of GLS collected in years of 2015 (batch 1 and 2), 2016 (batch 3, 4, and 5), 2017 (batch 6 and 7), and 2018 (batch 8, 9), respectively, and 14 batches of GLF collected in years of 2013 (batch 1), 2014 (batch 2), 2015 (batch 3, 4, and 5), 2016 (batches 6–11), 2017 (batch 12 and 13), and 2018 (batch 14), respectively.

In addition, 5 batches of GLSF (GLS was mixed with GLF with the ratio of 4:1, w/w, batch 1, 2, 3, 4, 5) were also analyzed.

Based on the literature search the main active components including ganoderenic Acid C (GNAC), ganoderic Acid C2 (GAC2), ganoderic Acid G (GAG), ganoderic Acid B (GAB), ganoderenic Acid B (GNAB), ganoderic Acid A (GAA), ganoderic Acid H (GAH), ganoderenic Acid D (GNAD), ganoderic Acid D (GAD), ganoderic Acid F (GAF), ganoderic Acid DM (GADM), ganoderol A (GLA), and ergosterol (EGL) were selected as the quantification markers (Fig. 1) and purchased from Chinese pure standards market with purity higher than 90% based on their corresponding HPLC chromatograms.

HPLC grade acetonitrile, methanol, and water were purchased from Fisher Scientific (Pittsburgh, PA, USA). Chloroform for extraction of herbs was purchased from BDH Chemicals (VWR International, Radnor, PA).

Methods

HPLC assay development and validation

The constitution of the mobile phase was firstly optimized. Because some of the active components are weak acids, phosphoric acid was used to make an acidic buffer (0.1%, v/v) to reduce the ionization of these compounds for better chromatographic resolution. Different elution conditions with acetonitrile and phosphoric acid (0.1%) were tested and compared. Because of the great difference in the polarity of these compounds, gradient elution was applied. The elution conditions were optimized which was able to separate these 13 compounds in 124 min (Table 3).

The column temperature was an important parameter influencing resolution and selectivity. Room temperature (around 25 °C), 30 °C and 35 °C were tested. The results indicated that 30 °C was the most suitable for the best resolution.

The selection of detection wavelength was essential for detecting the components of GL products, and thus it is important for establishing the HPLC fingerprint. Diode-array detector (DAD) was applied to select the optimal wavelength by testing four conditions including 210, 230, 254, and 360 nm. The results indicated that 254 nm were able to generate the most abundant peaks when compared to other wavelengths.

The representative chromatograms of LLOQ, LQC, MQC, and HQC samples are shown in Fig. 2A–D. The typical retention time is shown in Table 3. These results suggested that the thirteen compounds at high and low concentrations could be clearly detected. No interfering peaks at the retention time were observed for all the, which suggested a desirable specificity and selectivity.

The Rs results are shown in Table 3, which are all greater than 1.5, except for GAA and GAH, which is 1.1 (Table 3). Many approaches have been examined to improve the separation, including change of column, modification of the mobile phase and elution gradient, temperature, etc. None of them could further increase the Rs due to the extremely similar chemical structure of these two compounds. However, GAA and GAH were able to be resolved under current conditions with acceptable accuracy and precision. Therefore, we believe this method still be able to be used for quantification.

The calibration curve was found linear over concentrations of 0.3125–20 µg/mL for GNAC, GAA, GAH, and GNAD; 0.78125–50 µg/mL for GAC2, GAG, GAB, and GAF; 0.15625–10 µg/mL for GNAB, GAD, and GADM; 0.625–40 µg/mL for GRA; and 1.5625–100 µg/mL for EGL. The correlation coefficient (R ²) was calculated to be greater than 0.99 for these standard curves.

The lower limit of quantification (LLOQ) was 0.3125 µg/mL for GNAC, GAA, GAH, and GNAD; 0.78125 µg/mL for GAC2, GAG, GAB, and GAF; 0.15625 µg/mL for GNAB, GAD, and GADM; 0.625 µg/mL for GRA; and 1.5625 µg/mL for EGL.

The inter-batch accuracy and precision were determined by measuring five replicates of QC samples at four concentration levels (LLOQ, LQC, MQC, and HQC) on separate batches. The results are shown in Table 4. The intraday accuracy was within 87.0–114.6%, while the precision was within 11.96% for all the samples tested.

The stability of the 12 compounds after 3 cycles of freeze-thaw, 24 hours in the autosampler, and 48 hours at room temperature was tested (see Table 5). At high concentration (HQC), all the samples are stable with deviation from the control samples less than 15%.

The spiked sample recovery was ranged from 69 to 119%, 79 to 115%, and 85 to 105% for samples of GLS, GLF, and GLSF, respectively.

Quantification of the active components

The concentration of the thirteen compounds was determined using the standard curves, and their content in the herbal extracts was calculated. The results are shown in Tables 6 –8 and Fig. 3. Among the 13 standards, GLA was only detectable in the standard GLF sample with a very low quantity, but not in any other samples indicating that GLA may not be the main component in GLS or GLF samples (Tables 6 –8). It may only play a minor role in the activity of GL products.

EGL was found in all the GLS samples in a consistent amount (average of 0.9 mg/g with the relative standard deviation (RSD) less than 20%). However, EGL was very low in the standard GLS sample. The possible reason is the commercial products were processed samples while the standard GLS is the raw material without processing. So the extraction of EGL in the standard GLS may not be sufficient. In contrast, EGL showed a large variance of existence in GLF samples. It was only detectable in the standard GLF and 4 out of 14 batches of GLF samples (Table 7).

The results showed that the abundance of these compounds was much higher in the extractions of the fruiting body than that of spore samples, except EGL (GLS is 4.5 higher than GLF extracts, average 0.9 mg/g in GLS versus 0.2 mg/g in GLF). GNAC and GAC2 were not detectable in GLS, while the mean contents in GLF were 0.11 and 0.42 mg/g in GLF samples). The remaining compounds were identified with low amounts in the GLS samples (less than 0.02 mg/g), but substantial contents were detected in GLF extracts (0.07–1.17 mg/g).

Large variations were identified with RSD greater than 60% among the GLF samples (See Fig. 4), indicating significant variability in the chemical compositions. For the GLS extracts, only compounds of GADM and EGL were constantly detectable in these samples, while others could only be quantified in a few of these samples leading to sparse data available. Therefore, only the RSD of GADM and EGL were calculated, which are 65 and 18.9% for GADM and EGL, respectively.

The sample mixture of GLSF (GLS: GLF = 4:1, w/w) was also analyzed using the validated HPLC-DAD method. As expected, the amount of compounds identified was greater than GLS but lower than GLF. In addition, a large variation was observed for most of the compounds as well (Table 8).

HPLC fingerprint

The chromatographic fingerprints were generated for 14 batches of GLF, 9 batches of GLS, and their standard samples. Before analysis, the chromatograms were modified by excluding the solvent peak as well as the peaks after 60 min. Although the retention time for EGL was found ∼95 min, it was excluded for further data analysis. The retention time was found much longer compared with other peaks. Such a long time for elution tends to result in unstable retention time (e.g., 5% variation may cause more than 5-min difference), so it is not suitable for similarity analysis. In terms of specificity, EGL is a common component found in many yeasts and other fungal membranes (with a similar function to cholesterol) [20]. It is not unique for GL samples. Therefore it is not a typical chemical marker for GLS and GLF products. Therefore, only the Ganoderma acids and GLA were included for the fingerprint construction.

Two simulative median fingerprint chromatograms were then produced for GLS and GLF (without standard samples). The similarity values of each chromatograph against the corresponding simulative median fingerprint chromatograms, as well as their inter-batch difference, was calculated (Tables 9 and 10).

As shown in Fig. 5, the chemical profiles of 9 batches of GLS samples were not consistent. The similarity results showed these samples could be partitioned into 2 major groups (year 2015 and 2016 versus 2017 and 2018). It appeared that samples collected in 2015 and 2016 have similar quality (similarity values from 0.85–0.92), while samples in 2017 and 2018 are close to each other (0.96–0.99). The results indicated the quality of GLS is highly affected by the year of collection. All these TRT GLS products are not consistent with the standard GLS with similarity values less than 0.5.

The chromatograms of 14 batches of GLF are shown in Fig. 6. Similar to the GLS samples, the quality of GLF samples are not consistent according to the similarity analysis. The samples collected in the years 2013 and 2014 had similar chemical profiles (>0.7). The samples collected in years 2015 and 2016 exhibited similar quality as well except batch 11, which was considered to be an outlier since it was quite different from any other samples regardless of the year collected. The more recent samples (years of 2017 and 2018) shared similar profiles indicating similar quality.

The results showed that the collection time has a great impact on the chemical compositions of the GL products (both fruiting body and spore), although they were from the same sites. Therefore, monitoring the chemical profiles would be important to ensure quality consistency.

None of these commercial samples provided by TRT had a similar chemical composition as the standard GL spore or fruiting body samples. Since the commercial products recruited in this study are all processed samples, it is highly likely to have different chemical fingerprints as the standards.

In this study, a marketed product (GLSF) was also tested and showed similar profiles in chemical components with a similarity value greater than 0.77 (Fig. 7).

Principal component analysis

The PCA plots of GLS and GLF with their corresponding cluster dendrograms are shown in Fig. 8. For both GLS and GLF, the two PCs cumulatively explain 63.5% and 80.2% of the variance in the samples, which appeared to be sufficient to explain most of the variability in these products. The results are consistent with the similarity analysis, which also showed great variance in the chemical composition of GLF and GLS products. In general, the products harvested in similar years are more close to each other, although there are some outliers (such as GLF Batch 3 and 13; GLS Batch 5 and 7). HCA results showed that the quality of GLS and GLF products before 2017 are more closely related than that of after 2017. However, due to the small number of batches, the PCA and HCA results need to be further investigated.