Safety assessment of HEA‐enriched Cordyceps cicadae mycelia on the central nervous system (CNS), cardiovascular system, and respiratory system in ICR male mice

Abstract Cordyceps cicadae, an entomopathogenic fungus, is a source of traditional Chinese medicine in China. Due to the low yield of wild C. cicadae, artificial cultivation approaches will be needed to meet the increasing market demand. Using bioreactor culture can increase mass production and the abundance of the active component, N6‐(2‐hydroxyethyl)‐adenosine (HEA). Here, we describe a safety assessment for a novel mycelium preparation method. Many studies have confirmed the safety of C. cicadae mycelia. However, the acute safety pharmacology of the C. cicadae enriched with the high HEA (3.90 mg/g) compound has not been evaluated. This study evaluated the central nervous system (CNS), cardiovascular system, and respiratory system in ICR male mice via oral gavage administration. For each requested item, two batches of eight mice tested on a vehicle (0.5% carboxymethyl cellulose, CMC) and C. cicadae mycelia (1,000 mg/kg) were performed. The heart rate at 60 min for the vehicle and C. cicadae mycelium treatment was 700.3 ± 55.4 and 603.0 ± 42.3 bpm, respectively (p = .4279). For echocardiographic analysis, the LV mass of the vehicle and drug treatment was 86.7 ± 6.4 and 80.2 ± 7.7, respectively (p = .0933). In the respiratory test, the tidal volume of the vehicle and drug treatments was 0.11 ± 0.01 and 0.14 ± 0.01 at 60 min, respectively (p = .4262). These results demonstrate that the oral administration of HEA‐enriched C. cicadae mycelia is safe for the CNS, cardiovascular, and respiratory systems.

Previous studies have shown that C. cicadae can produce many active components, such as adenosine and N6-(2-hydroxyethyl)adenosine (HEA). HEA is an adenosine derivative; it is associated with control of the brain and coronary circulation and has antiinflammatory activity (Deng et al., 2020;Lu et al., 2015). Also, HEA has also been reported to have sedative-hypnotic activity Li et al., 2017). The yield of C. cicadae collected in the field is low and unstable and easily contaminated with soil mold before drying. Previous studies have shown that the functional composition and physiological activity of artificially cultured liquid-fermented C. cicadae mycelium has a similar effect as wild C. cicadae mycelium Hsu et al., 2015).
However, liquid fermentation may produce different metabolites due to different cultivation conditions. Thus, a safety evaluation is required. We have previously conducted safety assessments including 90-day subchronic toxicity (2000 mg/kg bw) , genotoxicity study, 28-day subacute toxicity study in LY pig (Jhou et al., 2016), prenatal developmental toxicity study (Li et al., 2017), and acute toxicity study , and no toxic symptoms have been observed.
In this study, we compare the HEA content from mycelium and various fruiting bodies and investigate the pharmacology safety of HEA-enriched C. cicadae mycelia (1,000 mg/kg bw). To our knowledge, no study has reported a HEA active compound content higher than reported here. This study aimed to investigate the safety of this high HEA content material on the cardiovascular, respiratory, and central nervous systems via oral administration.

| Materials and chemicals
The Cordyceps cicadae sample was collected from the mountainous region of New Taipei City in Taiwan. The sample was cultured on PDA (potato dextrose agar) for 14 days at 25℃ after strain isolation and identification by ITS sequencing (Li et al., 2019). Next, a mycelium agar block (1 cm 3 ) was transferred to a 2-L Erlenmeyer flask with 1.0 L PDB broth and cultured for 3 days on a rotary shaker at 120 rpm and for 25℃.
The fermented broth was inoculated into a 500-L fermenter (BioTop, Taichung, Taiwan) (composed of 1% soybean powder, 5% glucose, and 1% yeast extract) at 25℃, 60 rpm agitation, and 0.5 vvm aeration for 3 days. Then, the broth was scaled up to a 5-ton fermenter and cultured under the same growth parameters for another 5 days. The fermented HEA-enriched C. cicadae mycelia were harvested, then boiled at 100℃ for 1 hr, freeze-dried, and ground into a powder. Ultimately, the C. cicadae mycelium powder was mixed with the appropriate amount of distilled water before administration to mice.
Following the method described in Chen et al., (2015), the active compound HEA in C. cicadae mycelia was determined by a highperformance liquid chromatography equipped with an ultraviolet detector and a reverse-phase column (Luna 5μ C18(2), 250 × 4.6 mm; Phenomenex, Torrance, CA). The mobile phase contains 10 mmol/L KH 2 PO 4 and acetonitrile (94:6) with a 1.0 ml/min flow rate. The column was kept at 40℃.

| Animals
For each pharmacology safety study, the central nervous system, cardiovascular system, and respiratory system were tested in two batches: Each batch contains a total of 16 mice, with eight in the vehicle control groups and eight in the treatment group. A total of 48 CD1 (ICR) male mice (8-14 weeks old) were obtained and held at the Taiwan Mouse Clinic (Taipei, Taiwan). Different ages of mice were used in the following tests. The mice were allowed free access to a chow diet and reverse osmosis water for accommodation. The facility room environment is kept at 21 ± 2℃, 40%-70% humidity, and a 24-hr light-dark cycle (LD 12:12). The study protocol was approved by the Institutional Animal Care and Utilization Committee [IACUC number: 13-07-563 (approved on November 19, 2018)]. For the pharmacokinetic (PK) study, 6-week-old Sprague Dawley rats were obtained from BioLASCO Taiwan Co., Ltd. After quarantine and accommodation for 1 week, the rats were randomized and were applied in the PK study.

| CNS analysis
In brief, the mice were divided into two groups: vehicle control (0.5% CMC) or C. cicadae mycelia (1,000 mg/kg). CNS analysis of screening battery contains home-cage activity (1 hr) (11-week-old mice), Modified SHIRPA (11-week-old mice), and core body temperature (14-week-old mice). All mice were orally administered for 15 min before behavioral testing. Home-cage activity was recorded for 1 hr, and data were analyzed using Clever Sys HomeCageScan TM3.0. In Modified SHIRPA, the mice were settled in a viewing jar for 5 min. For the core body temperature measurement, the mice's temperature was measure using a rectal probe (KN-91; Natsume Seisakusho, Tokyo, Japan) before drug treatment (time =0) and 30 and 60 min after administration.

| Cardiovascular analyses
In this section, the same dosage concentrations were used as mentioned above, and 8-week-old mice were used. Telemetry electrocardiogram (ECG), blood pressure and pulse rate, and echocardiography were included in cardiovascular analyses. Telemetry ECG recording was conducted in their home cage for 15 min before drug treatment; each mouse was then transferred to a novel cage to test their cardiac function after administered via the oral route. Mice implanted with a DSI radiotelemetric transmitter (model TA11ETA-F10; DSI, St. Paul, MN, USA) were allowed to recover for 1 week. For each drug treatment, an ECG was recorded for 60 min and seven different time points (predrug, 5, 10, 15, 20, 30, and 60 min). The RR interval, heart rate, PR interval, P duration, QRS interval, QT interval, and corrected QT data were collected and analyzed.
For blood pressure analysis, all mice were placed in the channel mouse platform for 4-5 min to allow the temperature to stabilize at 38℃. Mice were then measured for blood pressure by tail-cuff plethysmography using a BP-2000 blood pressure analysis system (Visitech Systems Inc., Cary, North Carolina, USA) before drug treatment (time =0), and after drug treatment at 10-15 (15), 25-30 (30), and 55-60 (60) min.
For echocardiography, cardiac structure and function were assessed by high-frequency echocardiography using a Vevo 3,100 imaging system (VisualSonics, Toronto, Ontario, Canada). Mice were sedated using 2% isoflurane gas with oxygen for 5 min and placed on a prewarmed pad with continuous electrocardiographic monitoring.

| Respiratory analyses
The same grouping vehicle control (0.5% CMC) and C. cicadae mycelia (1,000 mg/kg) were used for the lung function analysis by mouse whole body plethysmographs (Buxco, Wilmington, North Carolina, USA). Before drug administration, mice were habituated to the Buxco chamber for 15 min; the respiratory activity was recorded for 5 min at the same time as a baseline. Subsequently, we measured the mice's lung functions for 1 hr after drug administration. Ten-weekold mice were used in this analysis.
The supernatant (400 μL) was evaporated to dryness under nitrogen, and the residue was reconstituted with an equal amount of 0.1% formic acid (in DDW). Next, 100 μL of each sample was spiked with 100 μL internal standard solutions (HEA concentration 1 mg/L) for LC-MS/MS analysis (Marlinge et al., 2017).

| Statistical analysis
Continuous data were analyzed by a general linear model, and data ranking was analyzed using a categorical modeling procedure. Fisher's exact test was used for data analysis of Modified SHIRPA, two-way F I G U R E 1 Time course of pH, biomass, residue glucose, HEA, and adenosine contents under 5-ton liquid fermentation of C. cicadae F I G U R E 2 HPLC analysis in C. cicadae mycelium and fruiting body. Patterns (a) and (b) were nymph and stroma of wild fruiting body, respectively. (c) The source from the artificial fruiting body and (d) C. cicadae mycelium produced from a 5-ton fermenter ANOVA was used for data analysis of core body temperature, blood pressure, lung function, and telemetry ECG, and two-tailed t test was used for data analysis of home-cage activity and echocardiography. In all cases, p-values <0.05 were considered significant.

| Growth curve of liquid-fermented C. cicadae
The growth curve of C. cicadae cultured in a 5-ton fermenter is shown in Figure 1. We recorded the growth interval from Day 1 to Day 5. The HEA concentration increased rapidly after Day 1, and the biomass reached the stationary phase at Day 3. After 5 days in culture, 2.10% biomass and 3.90 mg/g HEA content were reached.
For residue glucose (R-Glc) and adenosine, the concentrations start to decrease after Day 2; these two materials were identified as important factors for biomass growth and metabolism (Ke & Lee, 2019).
We observed that the adenosine concentration decreased during the fermentation; this phenomenon may be due to the conversion of HEA because HEA has been reported as an adenosine analog (Meng et al., 2015). HPLC was used to analyze the HEA concentration of liquid-fermented mycelia and artificial and wild fruiting bodies ( Figure 2). The HEA content of liquid-fermented C. cicadae mycelium was 3.90 mg/g, and that in wild and artificial fruiting bodies was 0.29 and 1.37 mg/g, respectively. The values for the other group's fruiting body and liquid fermentation were 1.10 and 1.36 mg/g, respectively (Ke & Lee, 2019). Compared with artificial and wild C. cicadae fruiting F I G U R E 3 Effects of vehicle (0.5% CMC) and C. cicadae mycelia (1,000 mg/ kg) on home-cage activity in ICR mice, (a) including distance traveled, (b) walking, (c) drinking, (d) feeding, (e) grooming, (f) hanging, (g) rearing up, (h) resting, (i) twitching, and (j) awakening. Data are presented as mean ± SEM (n = 8) bodies, liquid-fermented C. cicadae mycelium possesses the highest HEA value, and no research report has been found with a higher content than this.

| Safety assessment
According to Zhao et al., (2020), a test article in safety pharmacology should be used at or above the therapeutic dose to investigate the potential adverse effect. With regard to safety pharmacology studies, the most important evaluation includes central nervous systems, cardiovascular and respiratory, which functions with respect to lifesupporting functions (Abraham, 2010). The dosage in this study (C. cicadae mycelium 1,000 mg/kg ICR mice) that converts to a normal adult human having a body weight of 70 kg is 5.69 g per day, higher than the previous study dose (1.05 g/day/person) (Tsai et al., 2020).
To our knowledge, no study has reported the pharmacology safety profile of this high amount of HEA. To address this, in this study, we have investigated the safety assessment of C. cicadae mycelia on the central nervous, cardiovascular, and respiratory systems.

| CNS analyses
For home-cage activity analysis, a fully automatic analysis system was used to study unconstrained mouse behaviors, which included distance traveled, walking, drinking, feeding, grooming, hanging, rearing up, resting, twitching, and awakening. There were no significant differences between the two groups in all ten home-cage behaviors ( Figure 3). Moreover, we notice a trend in the C. cicadae mycelium group with lower food consumption, which may be due to their food consumption. C. cicadae mycelia have abundant fiber. Hanging and rearing up behaviors are related to intensity activity strength (Luby et al., 2012). The C. cicadae mycelium group has a trend with lower hanging and rearing up behaviors, but the results between test groups showed no statistical difference. Next, modified SHIRPA is a threestage protocol that was used to detect a wide range of behavioral, neurological, and physiological measures in the same group of animals (Zhang et al., 2016). In this study, it provides a preliminary analysis of the morphological and behavioral characteristics of mice. A total of 43 behavioral assessments of mice were recorded (Table 1). No significant differences were found in all behavior items between vehicle and C. cicadae mycelium group by Fisher's exact test (p = 1.0000).
For the core body temperature in ICR mice, the rectal tempera-  TA B L E 1 Modified SHIRPA in vehicle and C. cicadae mycelium treatment F I G U R E 4 Effects of core body temperature in ICR mice. Data are presented as mean ± SEM. versus. vehicle (0.5% CMC, n = 8; C. cicadae mycelia 1,000 mg/kg, n = 8) (Fiebig et al., 2018;Saegusa & Tab ata, 2003), we found no significant interaction effect and no significant treatment differences between the two groups by using two-way ANOVA analysis.

| Cardiovascular analyses
In this section, some cardiovascular function-related indicators, including telemetry ECG, blood pressure, and echocardiography, were investigated.

| Telemetry ECG
Normally, freely roaming mice have a heart rate (HR) of 550-725 beats per min (bpm) and a maximal HR of 725-815 bmp. In mice, the PR interval amounts to 30-56 ms, the QRS complex has a duration of 9-30 ms, and the corrected QT interval (QTc) ranges between 30 and 124 ms (Kaese et al., 2013). In our study, ECG recording was tested in two different groups, vehicle and C. cicadae mycelium treatment. The results showed that C. cicadae mycelium treatment had no obvious adverse effect on ECG. There were no significant changes observed in P duration (p =.2215), QRS interval (p =.8571), QT interval (p =.9536), and corrected QT (p =.7818) between both treatment groups ( Figure 5). Of note, in the telemetry ECG assays, mice showed a temporal increase in heart rate (and decrease in RR interval) due to a response to handling. Since each mouse was transferred to a novel cage to test their cardiac function after administered, this may explain a temporal rapid heart rate due to this transfer event. C. cicadae mycelium-treated mice demonstrated a fast recovery to their basal state levels on RR interval (p <.0001), heart rate (p <.0001), and PR interval (p =.0082) when compared to its vehicle after 20 min. This fast recovery could also come from the C. cicadae F I G U R E 5 Effects of telemetry ECG in ICR mice. ECG recording of (a) RR interval, (b) heart rate, (c) PR interval, (d) P duration, (e) QRS interval, (f) QT interval, and (g) corrected QT. Statistically, repeated-measures two-way ANOVA was used to compare between groups. The post-Bonferroni analysis presented in the figure of A, B and C, * and *p <.05, ** and **p <.01, and *** and ***p <.001 compared with basal level of predrug. Data are presented as mean ± SEM. (vehicle, n = 8; C. cicadae mycelia 1,000 mg/kg, n = 8) . HEA is a Ca 2+ antagonist; it may help to reduce the heart rate and depress contractility under physiological conditions in cardiovascular disorders (Furuya et al., 1983;Hsu et al., 2015). However, since this is a one dosage experiment, further long-term consumption will be needed to verify this phenomenon. Together, these data suggest C. cicadae mycelia has almost no effect on cardiac function.

| Blood pressure
For blood pressure, systolic and diastolic pressure and pulse were investigated. These parameters fluctuate within the normal physiological range. Compared with vehicle, the C. cicadae mycelium group was not significantly different in the systolic (p =.6422) and diastolic (p =.1283) blood pressure or pulse rate (p =.4279) responses ( Figure 6). The C. cicadae mycelium group had a slightly lower trend within the systolic and diastolic values and pulse rates. Although no significant difference was found, this decreasing trend may be due to the HEA contents of the C.
cicadae mycelium treatment. According to Furuya et al. (1983), HEA has been reported as a nature calcium antagonist and inotropic agent. And it has been reported that calcium antagonist can reduce cytosolic free calcium concentration vasodilation in the coronary beds and has lower blood pressure effects (Grossman & Messerli, 2004).

| Echocardiography
Echocardiography has been widely applied in determining cardiac functions and phenotypes in murine models. The technique is convenient and safe, and allows consecutive and repeated evaluation of cardiovascular physiological and pathologic characteristics in live animals (Wu et al., 2010). Global and regional left ventricular (LV) functions are well-known indicators of cardiac disease (Salm et al., 2006). The most commonly used indexes of global LV systolic function are fractional shortening (FS) and ejection fraction (EF) because they are easily measured (Chengode, 2016;Liu & Rigel, 2009).
FS and EF were positively correlated, and the higher value means a better heart contraction (Child et al., 1981;Yoshikawa et al., 2012).
LV mass is the weight of the left ventricle, which is a commonly used descriptor of cardiac status (Pickering, 2007).

| Respiratory analyses
Respiration is a vital function. Therefore, it is required to be assessed in the safety analysis. This is to check whether the test substance might affect airway function (Goineau et al., 2013).
In this study, the function examines the lung using noninvasive F I G U R E 6 Systolic (a) and diastolic (b) blood pressure and heart rate (c) in vehicle-and C. cicadae mycelium-treated mice were measured by a tail-cuff method. Data are mean ± SEM of eight mice (five recording sessions each) per group. bpm indicates beats per min vehicle-and C. cicadae mycelium-treated groups ( Figure 8).
Furthermore, we observed a trend in the C. cicadae mycelium group with higher tidal volume (TVb) and respiration resistance (Penh). This might be due to C. cicadae mycelium HEA contents, which appears to have a sedative function in pharmacological tests (Liu et al., 2017;Wang et al., 2013). According to Hitomi et al., (2019), an increase in TVb can have a positive impact on health, such as subjective stress-related feelings and regulating the autonomic nervous system.

| Pharmacokinetic study
To evaluate the test time used in this study, we conducted a pharma- This phenomenon shows that HEA is retained in the brain for at least over 60 min, and it can be metabolized completely after 1,440 min.
The dose of HEA by intravenous injection shows no toxicity signs were observed in rats during this pharmacology test.

CO N FLI C T O F I NTE R E S T
The authors declare that they do not have any conflict of interest.

E TH I C A L A PPROVA L
The study was carried out after the study's protocols and approved from the Institutional Animal Care and Utilization Committee [IACUC number: 13-07-563 (approved on November 19, 2018)].

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available on request from the corresponding author.