An Investigation of the Anti-Depressive Properties of Phenylpropanoids and Flavonoids in Hemerocallis citrina Baroni

The World Health Organization predicts that over the next several years, depression will become the most important mental health issue globally. Growing evidence shows that the flower buds of Hemerocallis citrina Baroni (H. citrina) possess antidepressant properties. In the search for new anti-depression drugs, a total of 15 phenylpropanoids and 22 flavonoids were isolated and identified based on spectral data (1D and 2D NMR, HR-ESI-MS, UV) from H. citrina. Among them, compound 8 was a novel compound, while compounds 1–4, 6, 9, 10, 15, 17, 24–26, 28, and 37 were isolated for the first time from Hemerocallis genus. To study the antidepressant activity of phenylpropanoids and flavonoids fractions from H. citrina, macroporous resin was used to enrich them under the guidance of UV characteristics. UHPLC-MS/MS was applied to identify the constituents of the enriched fractions. According to behavioral tests and biochemical analyses, it showed that phenylpropanoid and flavonoid fractions from H. citrina can improve the depressive-like mental state of chronic unpredictable mild stress (CUMS) rats. This might be accomplished by controlling the amounts of the inflammatory proteins IL-6, IL-1β, and TNF-α in the hippocampus as well as corticosterone in the serum. Thus, the monomer compounds were tested for their anti-neuroinflammatory activity and their structure–activity relationship was discussed in further detail.


Introduction
Hemerocallis citrina Baroni (H. citrina) belongs to Liliaceae family [1], which is distributed widely from Europe to Asia. H. citrina is utilized as food and medicine because it has a pleasant flavor and bioactive secondary metabolites [2]. The flower buds are harvested before the plant blooms and dried, and they are regarded as a useful meal. The benefits, such as improving sleep and curing depression, were initially recorded in the book Bencao Gangmu Shiyi. Research conducted in the 21st century indicates that H. citrina possesses a number of beneficial effects, including antidepressant properties [3], anti-inflammatory properties [4], alleviation of insomnia [4], improvement of hepatitis [5], and anticancer properties [6]. According to phytochemical studies, several classes of biologically active components are present, including alkaloids [7], flavones [8], terpenes [9], steroidal saponins [10], and phenolic glycosides [11]. The search for antidepressant active ingredients in natural medicines and foods has undergone a new strategy. There is evidence that the total phenols extract of H. citrina has antidepressant properties [12], but the exact active ingredients remain unknown. As a result, we conducted a systematic extraction chemical shift and coupling constants. Thus, the structure of compound 8 was determined as 3-O-(Z)-p-coumaroylquinide. The key HMBC correlations were shown in Figure 1.
To understand the role of phenylpropanoids and flavonoids of H. citrina, on the effect of anti-depression, the target compounds need to be enriched. A total of 70 ethanol elution fractions and 37 monomeric compounds were injected into HPLC system. The UV spectrum of each peak was studied. It showed that different kind of Different compounds exhibited different UV spectrum maximum absorption. (E)-p-coumaroylquinic acid and (Z)-p-coumaroylquinic acid exhibited absorption maxima at 310 and 306 nm, respectively. The UV spectrum maximum absorption wavelength of caffeoylquinic acid were around 218 and 326 nm. The UV spectrum maximum absorption wavelength of feruloylquinic acid were around 236 and 324 nm. The UV spectra of the compounds were shown in the supporting information. By analysis of the UV spectrum of the compounds from 70 ethanol elution fractions, 20%−3~30%−5 fractions were mixed and selected as H. citrina flower buds total phenylpropanoids extract (HFPE). The UV spectrum maximum absorption wavelength of quercetin and isorhamnetin glycosides were around 256 and 356 nm. The absorption maxima of kaempferol glycosides were around 266 and 348 nm. By analysis of the UV spectrum of the compounds from 70 ethanol elution fractions, 30%−6~70%−1 fractions were mixed and selected as H. citrina flower buds total flavonoids extract (HFFE). Its final yields were 0.43% (w/w) for HFPE and 0.61% (w/w) for HFFE compared with the crude drugs. The HFPE and HFFE were stored at 4°C before use. UHPLC-Q-TOF-MS was adopted to characterize the constituents in HFPE and HFFE ( Figure 3). The compounds were identified by comparing the MS data and retention times with that of the isolated compounds. At last, a total of 13 phenylpropanoids were confirmed from HFPE and a total of 21 flavonoids were identified from HFFE. Compounds 1, 6, and 30 were not detected for their low content. The results showed that the main constituents of HFPE were phenylpropanoids and the main constituents of HFFE were flavonoids. The content of each phenylpropanoid compound in HFPE and each flavonoid compound in HFFE were shown in Tables S1 and S2.
To understand the role of phenylpropanoids and flavonoids of H. citrina, on the effect of anti-depression, the target compounds need to be enriched. A total of 70 ethanol elution fractions and 37 monomeric compounds were injected into HPLC system. The UV spectrum of each peak was studied. It showed that different kind of Different compounds exhibited different UV spectrum maximum absorption. (E)-p-coumaroylquinic acid and (Z)p-coumaroylquinic acid exhibited absorption maxima at 310 and 306 nm, respectively. The UV spectrum maximum absorption wavelength of caffeoylquinic acid were around 218 and 326 nm. The UV spectrum maximum absorption wavelength of feruloylquinic acid were around 236 and 324 nm. The UV spectra of the compounds were shown in the supporting information. By analysis of the UV spectrum of the compounds from 70 ethanol elution fractions, 20%−3~30%−5 fractions were mixed and selected as H. citrina flower buds total phenylpropanoids extract (HFPE). The UV spectrum maximum absorption wavelength of quercetin and isorhamnetin glycosides were around 256 and 356 nm. The absorption maxima of kaempferol glycosides were around 266 and 348 nm. By analysis of the UV spectrum of the compounds from 70 ethanol elution fractions, 30%−6~70%−1 fractions were mixed and selected as H. citrina flower buds total flavonoids extract (HFFE). Its final yields were 0.43% (w/w) for HFPE and 0.61% (w/w) for HFFE compared with the crude drugs. The HFPE and HFFE were stored at 4 ℃ before use. UHPLC-Q-TOF-MS was adopted to characterize the constituents in HFPE and HFFE ( Figure 3). The compounds were identified by comparing the MS data and retention times with that of the isolated compounds. At last, a total of 13 phenylpropanoids were confirmed from HFPE and a total of 21 flavonoids were identified from HFFE. Compounds 1, 6, and 30 were not detected for their low content. The results showed that the main constituents of HFPE were phenylpropanoids and the main constituents of HFFE were flavonoids. The content of each phenylpropanoid compound in HFPE and each flavonoid compound in HFFE were shown in Tables S1 and S2.

Body Weight
As shown in Figure 4, the growth rate of CUMS group's body weight was significantly lower than that of control group (p < 0.01) at the end of the fifth week. However, it was reversed by feeding it 25 mg/kg each of HFPE and HFFE over a period of five weeks.

Body Weight
As shown in Figure 4, the growth rate of CUMS group's body weight was significantly lower than that of control group (p < 0.01) at the end of the fifth week. However, it was reversed by feeding it 25 mg/kg each of HFPE and HFFE over a period of five weeks. Administration of fluoxetine for 5 weeks also alleviate the reduction in body weight of CUMS rats. In this study, we found CUMS procedure caused slower weight gain, which was consistent with the literature [12,54]. It has been reported that chronic variable stress produced a decrease in body weight along the stress exposure. It may be related to physiological changes, anorexia, or an increase in basal corticosterone [55].
Administration of fluoxetine for 5 weeks also alleviate the reduction in body weight of CUMS rats. In this study, we found CUMS procedure caused slower weight gain, which was consistent with the literature [12,54]. It has been reported that chronic variable stress produced a decrease in body weight along the stress exposure. It may be related to physiological changes, anorexia, or an increase in basal corticosterone [55]. . Trends in weight growth rate in each group before and after CUMS treated. The values were represented as mean ± SD (n = 10). ** p < 0.01 compared with the control group, ## p < 0.01 compared with the model group.

Effects of HFPE and HFFE on Sucrose Preference Test (SPT)
Rodents naturally have a strong desire for sweets, however, when the rodents are in a model of CUMS, they are not predisposed to drink sucrose solutions. Therefore, detecting the degree of preference for sucrose solution can be used as a useful means to evaluate the symptoms of anhedonia and the degree of depression in animals [56,57]. As shown in Figure 5, before the CUMS technique, there was not much of a difference between the groups. Sucrose preference was significantly decreased than that of the control group, after 4 weeks of CUMS induced, which indicated that the model was established successfully. After administering fluoxetine, HPPE, and HPFE for five weeks, the fluoxetine group displayed significantly higher sucrose preference (82.40%, p < 0.01) in CUMS rats compared to the model group (64.44%). Similarly, oral administration of HFPE and HFFE resulted in significant restoration (p < 0.01) of sucrose preference to normal level in CUMS rats. The sucrose preference of the HFPE group was 79.59% and of the HFFE group was 81.47%. The anhedonia-like behavior that CUMS induced in SPT was clearly reversed with the use of HPPE or HPFE. . Trends in weight growth rate in each group before and after CUMS treated. The values were represented as mean ± SD (n = 10). ** p < 0.01 compared with the control group, ## p < 0.01 compared with the model group.

Effects of HFPE and HFFE on Sucrose Preference Test (SPT)
Rodents naturally have a strong desire for sweets, however, when the rodents are in a model of CUMS, they are not predisposed to drink sucrose solutions. Therefore, detecting the degree of preference for sucrose solution can be used as a useful means to evaluate the symptoms of anhedonia and the degree of depression in animals [56,57]. As shown in Figure 5, before the CUMS technique, there was not much of a difference between the groups. Sucrose preference was significantly decreased than that of the control group, after 4 weeks of CUMS induced, which indicated that the model was established successfully. After administering fluoxetine, HPPE, and HPFE for five weeks, the fluoxetine group displayed significantly higher sucrose preference (82.40%, p < 0.01) in CUMS rats compared to the model group (64.44%). Similarly, oral administration of HFPE and HFFE resulted in significant restoration (p < 0.01) of sucrose preference to normal level in CUMS rats. The sucrose preference of the HFPE group was 79.59% and of the HFFE group was 81.47%. The anhedonia-like behavior that CUMS induced in SPT was clearly reversed with the use of HPPE or HPFE.

Effects of HFPE and HFFE on Open-Field Test (OFT)
The OFT was conducted following the SPT. The crossing score (the count of the rats crossing lines) and rearing score (the count of the rats standing up) of rats during a 5 min test could indicate their locomotion activity. OFT was often used to evaluate locomotion, exploratory activity, and anxiety-like behaviors in new environmental conditions [58]. The crossing score could reflect locomotion activity of rats. It was shown that there was no significant difference in crossing score between five groups, which was consistent with discovered in the literature [12]. This indicated that HFPE and HFFE did not affect the spontaneous motor ability of rats.
The rearing score could reflect exploratory activity of rats. The rearing score of depressed rats decreased as a result of a decrease in curiosity or interest in exploring the external environment. The HFPE, HFFE, and fluoxetine treatments improved the rearing score of the CUMS rat, as shown in Figure 6B. The values were represented as mean ± SD (n = 10). **p < 0.01 compared with the control group, ## p < 0.01 compared with the model group.

Effects of HFPE and HFFE on Open-Field Test (OFT)
The OFT was conducted following the SPT. The crossing score (the count of the rats crossing lines) and rearing score (the count of the rats standing up) of rats during a 5 min test could indicate their locomotion activity. OFT was often used to evaluate locomotion, exploratory activity, and anxiety-like behaviors in new environmental conditions [58]. The crossing score could reflect locomotion activity of rats. It was shown that there was no significant difference in crossing score between five groups, which was consistent with discovered in the literature [12]. This indicated that HFPE and HFFE did not affect the spontaneous motor ability of rats.
The rearing score could reflect exploratory activity of rats. The rearing score of depressed rats decreased as a result of a decrease in curiosity or interest in exploring the external environment. The HFPE, HFFE, and fluoxetine treatments improved the rearing score of the CUMS rat, as shown in Figure 6B.   Effects of HFPE and HFFE on the sucrose preference before and after CUMS procedure. The values were represented as mean ± SD (n = 10). **p < 0.01 compared with the control group, ## p < 0.01 compared with the model group.

Effects of HFPE and HFFE on Open-Field Test (OFT)
The OFT was conducted following the SPT. The crossing score (the count of the rats crossing lines) and rearing score (the count of the rats standing up) of rats during a 5 min test could indicate their locomotion activity. OFT was often used to evaluate locomotion, exploratory activity, and anxiety-like behaviors in new environmental conditions [58]. The crossing score could reflect locomotion activity of rats. It was shown that there was no significant difference in crossing score between five groups, which was consistent with discovered in the literature [12]. This indicated that HFPE and HFFE did not affect the spontaneous motor ability of rats.
The rearing score could reflect exploratory activity of rats. The rearing score of depressed rats decreased as a result of a decrease in curiosity or interest in exploring the external environment. The HFPE, HFFE, and fluoxetine treatments improved the rearing score of the CUMS rat, as shown in Figure 6B. Effects of HFPE and HFFE on OFT test. (A) represents crossing score, (B) represents rearing score. Results were represented as mean ± SD (n = 10). ** p < 0.01 compared with the control group, # p < 0.05 and ## p < 0.01 compared with the model group. Figure 6. Effects of HFPE and HFFE on OFT test. (A) represents crossing score, (B) represents rearing score. Results were represented as mean ± SD (n = 10). ** p < 0.01 compared with the control group, # p < 0.05 and ## p < 0.01 compared with the model group.

Effects of HFPE and HFFE on the Forced Swimming Test (FST)
An immobile posture in the FST reflects a condition of helplessness or despair. FST was often used to assess depression-like behaviors in rats [59]. Figure 7 demonstrated that compared with control group, the immobility time of the CUMS model group was prolonged. The administrations of HFPE, HFFE, or fluoxetine reduced immobility during the FST compared with that in the CUMS model groups. In addition, the crossing score of OFT results revealed that no difference was observed in locomotion activity between the five groups, so HFPE and HFFE indeed have the ability to improve depressive symptoms. compared with control group, the immobility time of the CUMS model group was prolonged. The administrations of HFPE, HFFE, or fluoxetine reduced immobility during the FST compared with that in the CUMS model groups. In addition, the crossing score of OFT results revealed that no difference was observed in locomotion activity between the five groups, so HFPE and HFFE indeed have the ability to improve depressive symptoms.

Figure 7.
Effects of HFPE and HFFE on FST. Results were represented as mean ± SD (n = 10). * p < 0.05 compared with the control group, # p < 0.05 and ## p < 0.01 compared with the model group.

Effects of HFPE and HFFE on Serum Corticosterone (CORT) Level and the Inflammatory Level in Hippocampus
As Figure 8A shown, the serum CORT levels were significantly higher in the CUMS group compared to the control group (p < 0.01). However, when HFPE and HFFE were administered to CUMS rats, the blood CORT levels significantly decreased. By activating the glucocorticoid receptor and regulating the aberrant activity of the hypothalamus-pituitary-adrenal (HPA) axis, high levels of glucocorticoids may exacerbate depression symptoms [60,61]. Serum CORT is an indicator of depression in laboratory animals. It was shown that CUMS induced an obvious elevation of serum CORT levels in rats, which was consistent with the previous studies [62] and proved the validity of CUMS model. However, HFPE and HFFE treatment could significantly reduce the CORT level in CUMS rats, indicating the alleviation of depression severity.
The results also showed that the levels of IL-6, IL-1β, and TNF-α of the CUMS group were significantly increased compared with those in the control group. However, the levels of the aforementioned pro-inflammatory cytokines were obviously reduced after 5 weeks of therapy with HFPE, HFFE, or fluoxetine ( Figure 8). Numerous studies have Figure 7. Effects of HFPE and HFFE on FST. Results were represented as mean ± SD (n = 10). * p < 0.05 compared with the control group, # p < 0.05 and ## p < 0.01 compared with the model group.

Effects of HFPE and HFFE on Serum Corticosterone (CORT) Level and the Inflammatory Level in Hippocampus
As Figure 8A shown, the serum CORT levels were significantly higher in the CUMS group compared to the control group (p < 0.01). However, when HFPE and HFFE were administered to CUMS rats, the blood CORT levels significantly decreased. By activating the glucocorticoid receptor and regulating the aberrant activity of the hypothalamuspituitary-adrenal (HPA) axis, high levels of glucocorticoids may exacerbate depression symptoms [60,61]. Serum CORT is an indicator of depression in laboratory animals. It was shown that CUMS induced an obvious elevation of serum CORT levels in rats, which was consistent with the previous studies [62] and proved the validity of CUMS model. However, HFPE and HFFE treatment could significantly reduce the CORT level in CUMS rats, indicating the alleviation of depression severity.
The results also showed that the levels of IL-6, IL-1β, and TNF-α of the CUMS group were significantly increased compared with those in the control group. However, the levels of the aforementioned pro-inflammatory cytokines were obviously reduced after 5 weeks of therapy with HFPE, HFFE, or fluoxetine ( Figure 8). Numerous studies have shown that inflammation plays a significant role in depression. According to the inflammatory theory for depression, stress triggers inflammatory processes, which impair the body's ability to produce serotonin and regulate the HPA axis, resulting in depression [63]. Pro-inflammatory cytokines-including IL-6, IL-1β, and TNF-α-were increased in depressed patients' bodies [64]. In this study, CUMS significantly improved the hippocampal levels of IL-6, IL-1β, and TNF-α in rats, while HFPE and HFFE significantly reversed the increase in these pro-inflammatory cytokines. These results implied that the protective effect of HFPE and HFFE on rat behavior may be related to inhibiting the release of pro-inflammatory factors. [63]. Pro-inflammatory cytokines-including IL-6, IL-1β, and TNF-α-were increased in depressed patients' bodies [64]. In this study, CUMS significantly improved the hippocampal levels of IL-6, IL-1β, and TNF-α in rats, while HFPE and HFFE significantly reversed the increase in these pro-inflammatory cytokines. These results implied that the protective effect of HFPE and HFFE on rat behavior may be related to inhibiting the release of pro-inflammatory factors. Results were represented as mean ± SD (n = 8). ** p < 0.01 compared with the control group, # p < 0.05 and ## p < 0.01 compared with the model group.

Anti-Neuroinflammatory Activity and Structure-Activity Relationship
The anti-neuroinflammatory activity of H. citrina flower buds 80% EtOH extract (HFE), HFPE, HFFE, and the 37 isolated compounds were evaluated in LPS-induced BV2 microglial cells model by IC50 values of inhibiting NO production as shown in Table 2. The purity of the isolated compounds was more than 95% as calculated from their peak areas of HPLC. Based on the preliminary study, the IC50 values of HFPE and HFFE were significantly lower than those of HFE, indicating that anti-neuroinflammatory substances were enriched in HFPE and HFFE. It can be observed that the compounds 3, 4, 5, 7, 8, 13,  16, 24, 25, 28, 30, and 37 showed potential anti-neuroinflammatory effects with IC50 values Figure 8. Effects of HFPE and HFFE on serum CORT level and the inflammatory level in hippocampus of CUMS rats. ELISA for detecting CORT (A), IL-6 (B), IL-1β (C), and TNF-α (D) levels. Results were represented as mean ± SD (n = 8). ** p < 0.01 compared with the control group, # p < 0.05 and ## p < 0.01 compared with the model group.

Anti-Neuroinflammatory Activity and Structure-Activity Relationship
The anti-neuroinflammatory activity of H. citrina flower buds 80% EtOH extract (HFE), HFPE, HFFE, and the 37 isolated compounds were evaluated in LPS-induced BV2 microglial cells model by IC 50 values of inhibiting NO production as shown in Table 2. The purity of the isolated compounds was more than 95% as calculated from their peak areas of HPLC. Based on the preliminary study, the IC 50 values of HFPE and HFFE were significantly lower than those of HFE, indicating that anti-neuroinflammatory substances were enriched in HFPE and HFFE. It can be observed that the compounds 3, 4, 5, 7, 8, 13, 16, 24, 25, 28, 30, and 37 showed potential anti-neuroinflammatory effects with IC 50 values less than 100 µM, and compounds 8, 16, 24, 25, and 30 showed stronger inhibitory effects on NO production in LPS-induced BV2 cells in comparison with the positive drug indomethacin. Thus, this study allows a preliminary discussion of the structure-activity relationship about the role of quinic acid or quinide group of phenylpropanoids, the role of sugar moieties at C-3 of flavonoids, and the role of substituent group at C-3 of flavonoids.

Role of Quinic Acid or Quinide Group of Phenylpropanoids
In order to investigate the role of quinic acid or quinide group, the activities of compounds 1, 3, and 7 were compared. The results indicated that the introduction of quinic acid or quinide group to the coumaroyl increased the activity, as could also be seen from the activities of compounds 2, 4, and 8. Compound 3 (the coumaroyl linked to the C-3 hydroxyl of quinic acid) showed higher activity than compounds 9 and 11 whose connection sites were at C-4 and C-5. It also could be observed that compound 4 (connection site at C-3) had better activity than compounds 10 (connection site at C-4).

Role of Sugar Moieties at C-3 of Flavonoids
It was reported that flavonoid with a hydroxyl at C-3 exhibited a remarkable increase in anti-neuroinflammatory activity [65]. In our results, we compared the antineuroinflammatory activities of compounds 16-23, 24-29, and 30-37, it showed that the compounds (16, 24, and 30) with hydroxyl group at C-3 showed higher activity than compounds (17-23, 25-29, and 31-37) with sugar moieties at C-3. It concluded that the sugar moieties at C-3 would decrease the activity.

Role of Substituent Group at C-3 of Flavonoids
To explore the effect of substituent group at C-3 , the activities of compounds 16, 24, and 30 were tested. The results showed that compound 24 (IC 50 = 13.56 µM) with methoxyl at C-3 showed higher activity than compounds 16 (IC 50 = 17.48 µM) with hydroxy at C-3 and 30 (IC 50 = 21.99 µM) with hydrogen at C-3 . The results suggested that methoxyl or hydroxy at C-3 played a more important role in the activitythan hydroxy at C-3 .
Although the anti-neuroinflammatory activity of flavonoid glycosides is very weak, they may convert into corresponding aglycones (quercetin, kaempferol, and isorhamnetin) in vivo [66] to exert with strong anti-neuroinflammatory effects. These findings suggest that phenylpropanoids and flavonoids may play a potential inhibitory role in microglia-involved neuroinflammation, thus producing neuroprotective effects in inflammatoryrelated neuronal diseases including depressive disorder.

Plant Material
H. citrina flower buds were provided by Tiancheng Agricultural Development Co. Ltd., Qidong, Hunan, China. A voucher specimen (no. HF20200905) was identified by professor Ying Jia, and deposited at School of Traditional Chinese Materia Medica of Shenyang Pharmaceutical University.

Apparatus and Reagents
HPLC was applied on Agilent 1260 Infinity equipped with an Agilent G1365D multiwavelength detector with a pack of column (YMC-packed C 18

Nuclear Magnetic Resonance Spectrometry (NMR)
One-and two-dimensional NMR experiments were performed on a Bruker Ascend 600 NMR spectrometer (Bremen, Germany), in methanol-d 4 , using TMS as an internal standard. Chemical shifts are given in parts per million (δ) relative to the residual proton signals of the solvent (MeOH, δ H 3.31 and δ C 49.00) and coupling constants (J) are given in hertz (Hz).

Enrichment of HFPE and HFFE of H. citrina Flower Buds
The air-dried H. citrina flower buds (2.5 kg) were extracted with 25 L 80% EtOH by cold-dipping method (three times, 48 h each time). The extract was evaporated under reduced pressure to give a residue that was suspended in water. An appropriate amount of the solution was preserved as H. citrina flower buds extract (HFE) to test antineuroinflammatory activity. Then, the residual sample was extracted by the same amount n-butanol for three times. The n-butanol fraction was subjected to CC (AB macroporous resin; EtOH/H 2 O 0%, 10%, 20%, 30%, 50%, 70%, 90%) to afford 70 fractions (10 fractions for each elution solvent). These fractions were injected to 1260 Agilent HPLC system (Agilent, Santa Cara, CA, USA) with a DAD detector. Separation was carried out on a Phenomenex luna C 18 column (5 µm, 4.6 × 250 mm). Mobile phases were water with 0.1% formic acid (A) and methanol (B). Chromatographic separation was achieved using a gradient elution as follows: 0.01 min, 5%B; 0.01-60 min, 5% to 100% B. The injection volume of sample was 20 µL. The flow rate was 1 mL/min and the column temperature was 35 • C. The fractions were merged under the direction of UV spectra.

UHPLC-MS Detection and Data Analysis
The analyses were performed by UHPLC system (Shimadzu, Kyoto, Japan), which is equipped with a model of LC-30AD pump and a model of SIL-30AC autosampler. The water column (ACQUITY UPLC ® HSS T3 1.7 µm, 2.1 × 100 mm) was used for separation. Mobile phases were water with 0.1% of formic acid (A) (Ph = 3.02) and acetonitrile with 0.1% of formic acid (B). Chromatographic separation was achieved using a gradient elution as follows: 0.01-2 min, 2%; 2-16 min, 2% to 20% B; 16-19 min, 20% to 23% B; 19-30 min, 23% to 100% B; 30-33 min, 100% B; 33-33.1 min, 100% to 2% B; 33.1-35 min, 2% B. A sample volume of 4 µL was injected and introduced to the column with 0.4 mL·min −1 of the solvent flow rate. The column temperature was set at 35°C. MS analysis instrument with a Triple TOF 4600 system (SCIEX, Framingham, MA, USA) was performed in negative mode. The mass range was set at m/z 100-1000 Da. The ESI heater temperature was set at 500 • C. The IonSpray Voltage Floating was set at 4500 V. The collision energy and declustering potential energy were set at −10 and −100, respectively. Nebulizer gas, curtain gas, and auxiliary gas were set at 50, 35, and 50 psi. The information-dependent acquisition mode was used for MS/MS ion data acquired. The MS conditions were corrected by APCI negative calibration solution for the AB SCIEX Triple TOF TM systems. PeakView software (version 2.2, AB SCIEX, CA, USA) was used for structural identification of compounds from H. citrina.

Animals
The Animal Ethics Committee of Shenyang Pharmaceutical University has approved all animal testing (no. SYPU-IACUC-S2021-03-03-202). Male Sprague-Dawley (SD) rats weighing 180-200 g were supplied by Central Animal House of Shenyang Pharmaceutical University (Shenyang, China). Animals were housed in a room that had a 12 h light/dark cycle and a temperature range of 21 to 25 • C. They were also provided water and a regular food. Before the actual trials start, the rats should acclimate to their new environment for seven days.

Groups and Drug Administration
A baseline test was conducted as described below before grouping. Rats were divided into five groups of 10 depending on their weight and sucrose preference. The model group, fluoxetine (10 mg/kg) group, HFPE (25 mg/kg) group, and HFFE (25 mg/kg) group were the four treatment groups using CUMS procedures. An untouchable control group was used. Based on what had been reported in the pharmacological literature [2,12] the HFPE and HFFE dosages were selected. Two hours before to the daily CUMS procedure and behavioral testing, the rats received the appropriate solutions. All of the medications were administered through gastric gavage at a volume of 3 mL/kg of body weight and dissolved in sodium carboxymethylcellulose (CMC-Na) at a concentration of 0.3%. The FST, OFT, and SPT were all conducted following the CUMS procedure, as shown in Figure 9. Days 76-79, 80, and 81 after the final dosage of the medication, respectively, saw the completion of the SPT, OFT, and FST. Following the final test, blood samples were collected and centrifuged at 4000 rpm for 10 min. Rat serum and brain tissue were collected and stored at −80 • C for future study.
with a Triple TOF 4600 system (SCIEX, Framingham, MA, USA) was performed in nega-tive mode. The mass range was set at m/z 100-1000 Da. The ESI heater temperature was set at 500 °C. The IonSpray Voltage Floating was set at 4500 V. The collision energy and declustering potential energy were set at −10 and −100, respectively. Nebulizer gas, curtain gas, and auxiliary gas were set at 50, 35, and 50 psi. The information-dependent acquisition mode was used for MS/MS ion data acquired. The MS conditions were corrected by APCI negative calibration solution for the AB SCIEX Triple TOF TM systems. PeakView software (version 2.2, AB SCIEX, CA, USA) was used for structural identification of compounds from H. citrina.

Animals
The Animal Ethics Committee of Shenyang Pharmaceutical University has approved all animal testing (no. SYPU-IACUC-S2021-03-03-202). Male Sprague-Dawley (SD) rats weighing 180-200 g were supplied by Central Animal House of Shenyang Pharmaceutical University (Shenyang, China). Animals were housed in a room that had a 12 h light/dark cycle and a temperature range of 21 to 25 °C. They were also provided water and a regular food. Before the actual trials start, the rats should acclimate to their new environment for seven days.

Groups and Drug Administration
A baseline test was conducted as described below before grouping. Rats were divided into five groups of 10 depending on their weight and sucrose preference. The model group, fluoxetine (10 mg/kg) group, HFPE (25 mg/kg) group, and HFFE (25 mg/kg) group were the four treatment groups using CUMS procedures. An untouchable control group was used. Based on what had been reported in the pharmacological literature [2,12] the HFPE and HFFE dosages were selected. Two hours before to the daily CUMS procedure and behavioral testing, the rats received the appropriate solutions. All of the medications were administered through gastric gavage at a volume of 3 mL/kg of body weight and dissolved in sodium carboxymethylcellulose (CMC-Na) at a concentration of 0.3%. The FST, OFT, and SPT were all conducted following the CUMS procedure, as shown in Figure  9. Days 76-79, 80, and 81 after the final dosage of the medication, respectively, saw the completion of the SPT, OFT, and FST. Following the final test, blood samples were collected and centrifuged at 4000 rpm for 10 min. Rat serum and brain tissue were collected and stored at −80 °C for future study.

Body Weight and CUMS Procedure
Body weights were recorded every week. The CUMS procedure was carried out as Willner [67] described with a minor adjustment. CUMS referred to exposure to a variety

Body Weight and CUMS Procedure
Body weights were recorded every week. The CUMS procedure was carried out as Willner [67] described with a minor adjustment. CUMS referred to exposure to a variety of different variable stress factors, including: (1) being deprived of food for 24 h, (2) being deprived of water for 24 h, (3) 45 • cage tilt for 24 h, (4) being in an empty cage, (5) the light/dark cycle of inversion, (6) 5 min cold swimming (4 • C), (7) 24 h in wet sawdust, (8) horizontal cage shaking for 30 min, (9) 1 min tail nipping, and (10) 4 h physical restraint. The aforementioned stressors were distributed at random to each of the other four rat groups during the course of the experiment's nine weeks, but not to the blank group.

Sucrose Preference Testing
The SPT is used to evaluate rodent behavior associated with a human clinical symptom of depression by evaluating the capacity to seek pleasure. SPT was carried out in the same manner as previously described [68]. After one week of adapted feeding, the SPT were performed at 4 weeks' CUMS exposure and the last CUMS exposure. Each rat was given two pre-weighed bottles of either water or a solution containing 1% sucrose (w/v) for 6 h after being without water for 12 h. The bottles were replaced for an additional 6 h. After that, the amount of liquids consumed was recorded. The method used to determine sucrose preference is as follows: consumption of sucrose divided by consumption of sucrose plus water equals sucrose preference.

Open-Field Test
To assess the effects of HFPE and HFFE on exploratory behavior of rats administered CUMS, OFT carried out in the same manner as literature [69]. Each rat was placed in its own square in the center of a black box that measured 100 cm × 100 cm × 40 cm. There were 25 squares on the floor. For five min, the rats were free to walk around. Both the crossing score and the rearing score for each animal were recorded. The apparatus was carefully cleaned with 75% ethanol and rinsed to remove any odors.

Forced Swimming Test
The FST protocol was carried out as described in previous literature [70]. Each rat was placed in a separate transparent bucket filled with 23 ± 2 • C water that was 70 cm high and 40 cm wide. The time the rat's immobility duration in the last five minutes of their six-minute swim in the bucket was recorded. The apparatus was cleaned after each usage, the rats were dried and returned to their cages.

Determination of Serum CORT Level and IL-6, IL-1β, and TNF-α Level in the Hippocampus
The serum and hippocampal tissue of rats were taken. The tissue was appropriately diluted before the reagents and samples were put one at a time to the microtiter plate in accordance with the ELISA kit's instructions. The optical density value was calculated after the reaction by setting the microtiter plate at 450 nm. The amount of CORT in the serum and the amounts of IL-6, IL-1β, and TNF-α in the hippocampus tissue were calculated using the standard curve.
3.8. Anti-Neuroinflammatory Activity 3.8.1. Cellular Culture BV2 microglia cells were purchased from iCell Bioscience Inc. (Shanghai, China). The cells were grown at 37 • C in a humid environment with 5% CO 2 . In DMEM with 10% fetal bovine serum and 1% penicillin-streptomycin solution, they were maintained alive. The following assays were carried out while the cells were in the exponential growth phase (subcultured 2-3 times).

CCK8 Cytotoxic Activity
The cytotoxic activity was assessed using the CCK8 test. In a 96-well plate, BV2 microglial cells (5 × 10 4 cells/well) were placed in each well and incubated for 12 h. After one hour with each substance, LPS (1 µg/mL) was applied to the BV2 microglia cells for an additional 24 h. The CCK8 solution was then incubated with BV2 cells for 1 h at 37 • C. The absorbance at 450 nm was measured using a plate reader (Molecular Devices, San Jose, CA, USA). The cells that had not been treated were assumed to have a 100% optical density.

Inhibition of NO Production
To determine how much NO is produced, the literature [18] was consulted. Equal parts of the Griess reagent (1% sulfanilamide and 0.1% N-(1-naphthyl) ethylenediamine dihydrochloride in 2.5% phosphoric acid) were combined with the sample-treated cell culture medium for 5 min at room temperature and in the dark. It was determined that the absorbance was at 540 nm using a microplate reader. The trials were conducted simultaneously in triplicate.

Statistical Analysis
All data were expressed as the means ± SD from three independent experiments and were analyzed using GraphPad Prism 8.0 software (GraphPad software, San Diego, CA, USA). Statistical analysis was performed by one-way ANOVA with Tukey's post-hoc test, and p < 0.05 was regarded as significant difference.

Conclusions
In the present study, we describe the isolation and structure identification of bioactive phenylpropanoids and flavonoids from H. citrina together with an evaluation of their anti-neuroinflammatory activity. As part of our ongoing interest in discovering active ingredients from natural products, 15 phenylpropanoids and 22 flavonoids were isolated and identified from H. citrina. Among them, compound 8 was a novel compound, compounds 1-4, 6, 9, 10, 15, 17, 24-26, 28, and 37 were isolated from Hemerocallis genus for the first time.
Importantly, HFPE and HFFE were successfully enriched, and a total of 13 phenylpropanoids were confirmed from HFPE and a total of 21 flavonoids were confirmed from HFFE. According to the present SPT, OFT, and FST studies, HFPE and HFFE can improve depression-like behavior in rats. The biochemistry analyses of serum CORT level and the IL-6, IL-1β, and TNF-α level in hippocampus of CUMS rats indicate that the protective effect of HFPE and HFFE on rat behaviors may be associated with the release of CORT and pro-inflammatory cytokine.

Informed Consent Statement: Not applicable.
Data Availability Statement: Not available.