Chemicals and reagents

Cordycepin (>98%) was purchased from Winherb (Shanghai, China). Isocordycepin (99.87%) was purchased from Aladdin (Shanghai, China). Acetic acid (LC-MS grade) was obtained from Aladdin (Shanghai, China). Ethanol (HPLC grade) was obtained from Energy-chemical (Shanghai, China). The deionized water was purified by Milli-Q purification system (Millipore, USA).

Materials

40 batches of Cordyceps samples (10 different kinds of Cordyceps, Fig. 1), including 10 batches of cultured C. sinensis, 10 batches of natural C. sinensis, 3 batches of H. sinensis, 3 batches of C. militaris fruiting body, 3 batches of C. militaris, 3 batches of C. chanhua, 2 batches of Cordyceps haukesii, 2 batches of Cordyceps liangshanensis, 2 batches of Cordyceps gracilis and 2 batches of Paecilomyces hepiali were collected from different areas of China. The sample information was listed in (Table 1). The authenticity of all samples were identified by Dr Qian. Voucher specimens were deposited at the Key Laboratory of State Administration of Traditional Chinese Medicine, Dongguan, Guangdong.

View Full Size

The pictures of Cordyceps samples

Citation: Acta Chromatographica 36, 1; 10.1556/1326.2022.01094

• Download Figure
• Download figure as PowerPoint slide

Preparation of standard solutions

The stock standard solution of cordycepin (499.3 μg mL⁻¹) and isocordycepin (509.1 μg mL⁻¹) was prepared with deionized water and further diluted with deionized water to obtain a series of working solutions. The standard solutions were kept at 4 °C.

Preparation of sample solution

The sample was homogeneously powdered in a mill, passed through a 65 mesh sieve, and then extracted by boiling water extraction (BWE). The following procedure was modified from a literature [19]. 0.1 g powdered sample and 10.0 mL boiling water (95 °C–97 °C) were added into a conical flask. The mixture was extracted ultrasonically (33 kHz, 250 W) at 75 °C for 30 min, and then cooled at room temperature; water was subsequently added to compensate for the lost weight. The final solution was filtered through a 0.22 µm membrane prior to LC-MS/MS analysis.

LC-MS/MS condition

The analysis was performed on an Agilent 1290 Infinity UPLC system (Agilent Technologies, USA) equipped with a binary pump, an auto-sampler, a column oven, and a Triple Quadrupole 6470 mass spectrometer (QQQ). The LC separation was performed on an Agilent poroshell 120 SB-Aq C18 column (3.0 × 50 mm, 2.7 µm) with 2% ethanol containing 0.2% acetic acid as the mobile phase at the flow rate of 0.6 mL min⁻¹. The column temperature was maintained at 30 °C. The injection volume was 2 µL.

QQQ equipped with a Jet Stream electrospray ionization (ESI) source was operated in positive ionization mode with “multiple reaction monitoring” (MRM). MRM transition (252→136) (fragmentor voltage 70 V; collision energy 20 V) was employed for quantitative analysis and MRM transition (252→119) (fragmentor voltage 70 V; collision energy 40 V) was employed for validation; Other conditions were as follows: drying gas (N2) flow rate 10 L/min; drying gas temperature 350 °C; sheath gas flow 11 L min; sheath gas temperature 350 °C; nebulize 35 psi; capillary voltage 4.0 kV; dwell time 500 ms.

Method validation

The current method was validated for linearity, limit of quantification (LOQ), limit of detection (LOD), precision, accuracy, and stability according to Chinese Pharmacopoeia [20].

Optimization of extraction condition

Various extraction methods for analyzing nucleosides in Cordyceps have been reported [21, 22]. Ultrasonic extraction was the most common one. To be noted, different types of extraction solvent resulted in different extraction contents of nucleosides in the previous studies [23, 24]. Qian et al. reported that the adenosine content of C. sinensis extracted by ambient temperature water extraction (ATWE) was significantly higher than that extracted by boiling water extraction (BWE), and further study proved that adenosine triphosphate (ATP)/adenosine diphosphate (ADP)/adenosine monophosphate (AMP) could be transformed into adenosine, and adenosine could be transformed into inosine at ATWE condition [24]. Tian et al. reported that the content of isocordycepin in C. sinensis extracted by water was higher than that extracted by 80% methanol, which indicated that transformation might also exist in cordycepin and isocordycepin [12].

To obtain the suitable extraction condition of cordycepin and isocordycepin, the comparison of ATWE and BWE was carried out. Deoxyadenosine triphosphate (dATP), deoxyadenosine diphosphate (dADP), deoxyadenosine monophosphate (dAMP), isocordycepin, cordycepin and 2′-deoxyinosine were monitored. The comparison results of ATWE and BWE were listed in the supporting information. As shown in Figure S1, only dAMP and isocordycepin were detected in the two extraction solutions. The content of dAMP in ATWE solution was lower than BWE solution, while the content of isocordycepin in ATWE solution was higher than BWE solution. The results proved that in ATWE condition, dAMP would be transformed into isocordycepin, while the transformation could be inhibited in BWE condition. Thus, the BWE condition was used for the extraction of free cordycepin and isocordycepin from samples.

Optimization of LC-MS/MS condition

The cordycepin and isocordycepin, which were usually separated on a reverse phase column with a high ratio of water, were polar components [12, 25, 26]. In the literature, the LC separation of cordycepin and isocordycepin was operated on an Agilent ZORBAX SB-Aq (4.6 × 150 mm, 5 μm) column, and 15 min were consumed [12]. Poroshell column, which consists of silica particles with a fused core and a layer of porous silica coating, can provide rapid separation of compounds with low back pressure. To reduce the separation time and solvent consumption, a smaller internal diameter and smaller particle size Agilent poroshell SB-Aq C18 column (3.0 × 50 mm, 2.7 μm) was utilized. To avoid the harmful organic solvent consumption, ethanol was selected as the organic mobile phase. Comparing 1%, 2%, and 4% ethanol, the 2% ethanol condition was chosen as the optimal condition because of the suitable compounds resolution and separation time. Two mobile phase additives (formic acid and acetic acid) were also compared. Acetic acid-ethanol system was selected for the better resolution. The flow rates of 0.5, 0.6, and 0.7 mL min⁻¹ were tested, 0.6 mL min⁻¹ was used for the better separation and shorter time. In addition, the different column temperatures (25, 30, and 35 °C) were compared, and the results showed that the higher column temperature could reduce the separation time, but too high temperature would decrease the resolution. Eventually, 30 °C was selected.

In the current study, the MRM mode of LC-MS/MS was used for detection two analytes. Cordycepin and isocordycepin could be easily ionized into protonated precursor ions ([M+H] ⁺), thus the positive mode was used. After elimination of ribodesose residues, the precursor ions (m/z 252) of cordycepin and isocordycepin produced diagnostic product ions (DPIs) associated with adenine (m/z 136) (Fig. 2), MRM transition (252→136) was picked as the quantitative ions. To obtain a good sensitivity, the collision energy (40, 30, 20, 10 V) was optimized. According to the relative response abundance, 20 V was selected. To ensure the peak was correct, the second intensity of DPIs (m/z 119), which was produced by elimination of NH3 from adenine (Fig. 2), was picked for validation; the collision energy was optimized and 40 V was selected.

View Full Size

The MS spectra and fragmentation pathways for cordycepin (A) and isocordycepin (B)

Citation: Acta Chromatographica 36, 1; 10.1556/1326.2022.01094

• Download Figure
• Download figure as PowerPoint slide

Method validation

The calibration curve of cordycepin was y = 7904.17x−4078.36 (r = 0.9996) within the tested range (0.9986–99.86 ng ml⁻¹), of isocordycepin was y = 10351.58x−6918.81 (r = 0.9994) within tested range (1.018–101.8 ng ml⁻¹), which indicated good linearity of the two compounds. The LODs for cordycepin and isocordycepin were 0.040 and 0.041 pg. The LOQs for cordycepin and isocordycepin were both 0.10 pg.

The RSDs of intra- and inter-day precision (n = 6) for cordycepin were 2.1% and 3.4%; The RSDs of intra- and inter-day precision (n = 6) for isocordycepin were 3.8% and 3.2%. The recoveries of cordycepin and isocordycepin (n = 6) were 101.2–105.4% (RSD = 1.6%), 94.4–109.5% (RSD = 5.5%), respectively. The RSDs of stability (24 h) for cordycepin and isocordycepin were 2.9% and 3.6%. The developed LC-MS method was sensitive and accurate for the determination of cordycepin and isocordycepin in Cordyceps samples.

Analysis of samples

The validated method was utilized to determine the contents of cordycepin and isocordcepin in different Cordyceps samples. The representative chromatograms of different Cordyceps were shown in (Fig. 3) and the results were listed in (Table 1).

View Full Size

The Chromatograms of Cordyceps samples: A. blank; B. STD; C. Cultured Cordyceps sinensis; D. Natural C. sinensis; E. Hirsutella sinensis; F. Cordyceps militaris fruiting body; G. Cordyceps militaris; H. Cordyceps chanhua; I. Cordyceps haukesii; J. Cordyceps liangshanensis; K. Cordyceps gracilis; L. Paecilomyces hepiali. 1. Cordycepin, 2. Isocordycepin

Citation: Acta Chromatographica 36, 1; 10.1556/1326.2022.01094

• Download Figure
• Download figure as PowerPoint slide

Cordycepin was only detected in Cordyceps millitaris and C. millitaris fruiting body samples. Isocordycepin was detected in C. sinensis, H. sinensis, C. chanhua, C. haukesii, C. liangshanensis, C. gracilis and Paecilomyces hepialid samples, while not detected in C. millitaris and C. millitaris fruiting body samples. This result was in line with the recent genome and chemical reports: C. millitaris could produce the cordycepin [9], while C. sinensis contain the isocordycepin [11, 12].

The contents of isocordycepin in cultured and natural C. sinensis were ranged from 0.43 to 2.69 μg g⁻¹ and from 0.67 to 1.40 μg g⁻¹, respectively. T test was applied to specify the difference between the content of isocordycepin in cultured and natural C. sinensis. P-value was obtained as 0.11, which revealed that there was no significant difference. Compared with previous studies, the contents of isocordycepin in C. sinensis of present experiment were lower [12, 13]. The reason may be that the contents of isocordycepin in previously reported methods included the free and transformed isocordycepin, while the present method only reflected the free isocordycepin. The contents of isocordycepin in H. sinensis, C. gracilis, and P. hepialid were similar to C. sinensis. The contents of isocordycepin in C. chanhua, C. haukesii, C. liangshanensis were slightly higher than C. sinensis. Interestingly, the contents of cordycepin in C. millitaris and C. millitaris fruiting body were significantly higher than the contents of isocordycepin in other Cordyceps samples. In addition, the contents of cordycepin in C. millitaris was higher than that in C. millitaris fruiting body, which might be caused by that cordycepin might be prone to be produced in the caterpillar than solid medium.

Comparison with reported LC methods

A comparison of the proposed method and the two reported methods for the determination of codycepin and isocordycepin in Cordyceps was exhibited in (Table 2). In terms of simplicity, the proposed method was much simpler than method 1 and method 2, because method 1 and method 2 needed extra online SPE and derivatization, respectively. In terms of analytical time, the proposed method was the fastest, which only consumed 35 min, including 30 min of sample extraction and 5 min of LC-MS separation. While method 1 consumed 1,695 min in total, including 1,680 min of sample preparation (extraction and derivatization) and 15 min HPLC-FD separation; method 2 took 105 min, including 90 min of sample extraction and 15 min of online SPE-HPLC separation. In terms of sensitivity, the proposed method was the best with LOD of 0.04 pg. Compared with method 2, the sensitivity was remarkably improved 12,300 folds; compared with method 1, the sensitivity was also improved 5 folds. In terms of eco-friendly, the developed method only consumed 6 μL acetic acid and 60 μL ethanol, while methods 1 and 2 both used harmful organic solvents, such as acetonitrile and methanol. In a word, the proposed method was a simple, rapid, sensitive, and eco-friendly method for the determination of cordycepin and isocordycepin in Cordyceps samples.