Circadian Gene PER2 Silencing Downregulates PPARG and SREBF1 and Suppresses Lipid Synthesis in Bovine Mammary Epithelial Cells

Simple Summary The present study was constructed to determine the effects of the core circadian clock gene, Period 2 (PER2), on lipid synthesis in bovine mammary epithelial cells (BMECs). Data revealed that PER2-regulated genes were involved in fatty acid de novo synthesis, desaturation, TAG accumulation, and lipid droplet secretion in primary BMECs, partly by inhibiting PPARG and SREBF1. Our overall data suggests that PER2 in bovine mammary cells plays a role in regulating milk fat synthesis directly, or via the activation of the transcription regulators PPARG and SREBF1. This study provides molecular evidence underscoring a link between the circadian clock and lipid metabolism in bovines. Abstract PER2, a circadian clock gene, is associated with mammary gland development and lipid synthesis in rodents, partly via regulating peroxisome proliferator-activated receptor gamma (PPARG). Whether such a type of molecular link existed in bovines was unclear. We hypothesized that PER2 was associated with lipid metabolism and regulated cell cycles and apoptosis in bovine mammary epithelial cells (BMECs). To test this hypothesis, BMECs isolated from three mid-lactation (average 110 d postpartum) cows were used. The transient transfection of small interfering RNA (siRNA) was used to inhibit PER2 transcription in primary BMECs. The silencing of PER2 led to lower concentrations of cellular lipid droplets and triacylglycerol along with the downregulation of lipogenic-related genes such as ACACA, FASN, LPIN1, and SCD, suggesting an overall inhibition of lipogenesis and desaturation. The downregulation of PPARG and SREBF1 in response to PER2 silencing underscored the importance of circadian clock signaling and the transcriptional regulation of lipogenesis. Although the proliferation of BMECs was not influenced by PER2 silencing, the number of cells in the G2/GM phase was upregulated. PER2 silencing did not affect cell apoptosis. Overall, the data provided evidence that PER2 participated in the coordination of mammary lipid metabolism and was potentially a component of the control of lipid droplets and TAG synthesis in ruminant mammary cells. The present data suggested that such an effect could occur through direct effects on transcriptional regulators.


Introduction
In non-ruminants, peripheral clocks exhibit a rhythmic oscillation with autonomic rhythms in almost all peripheral tissues and organs, including the mammary gland [1][2][3].
In the bovine mammary gland, there is also evidence that a number of circadian rhythm genes, some of which have well-known metabolic roles, play a key role in mammary gland function [4,5].
In cows, indirect evidence indicates that endogenous circadian clock systems could represent important control points of the biological rhythms associated with physiology and biochemistry. For example, rectal temperatures, respiratory rates, plasma concentrations of metabolites (glucose, urea, and cholesterol) [6], and concentrations of the growth hormones [7] prolactin [8] and cortisone [9] exhibit circadian oscillations in lactating cows. The synthesis of milk components also appears to follow a rhythmic pattern, which can be regulated, in part, by a circadian mechanism [10][11][12].
In rodents, Period circadian regulator 2 (PER2), a core circadian clock gene, plays a role in lipid metabolism. PER1/2-deficient mice have reduced liver TG levels [13], suggesting that PER could be involved in lipid metabolism. Another study in a rat model revealed that PER2 controls lipid metabolism via targeting PPARG expression [14]. Thus, PER2 might represent a novel control point in the bovine mammary lipid metabolism. The PER2 gene is not only associated with lipid metabolism, but it also shares inextricable molecular links with cell cycle oscillators [15,16]. In non-ruminants, the synchronized physiological signals of the circadian clock could lead to cell aggregation and division [17], indicating that the cell cycle is synchronized by the circadian clock. Although data supports an important role for PER2 in lipid metabolism, such effects, and the molecular mechanisms in bovine mammary epithelial cells (BMECs), are largely unknown.
In the present study, we hypothesized that PER2 regulated lipid metabolism and was potentially associated with cell cycle control in BMECs. To test this hypothesis, small interfering RNA (siRNA) was used to silence PER2 expression in primary BMECs. We investigated the role of PER2 on cell cycle activity and lipid synthesis at a transcription level, aiming to provide molecular evidence to support the link between the circadian clock and lipid metabolism in the bovine.

Cell Transfection
Small interfering RNA (siRNA) was used to inhibit PER2 expression in primary BMECs. The primary BMECs were transfected with three siRNAs, targeting the PER2 gene at 0, 12, 24, 36, 48, and 60 h, to screen for optimal siRNA and its transfection time point. The real-time PCR was employed to determine the transfection efficiency.
The primary BMECs were divided into a PER2 silencing group and a non-transfected group (without transfection, blank control). Negative siRNA was used as negative control during the real-time PCR. SiRNA and negative siRNA was synthesized by GenePharma Co., Ltd., (Genepharma Co., Ltd, Shanghai, China) and the sequences are listed in Table S1. The transfection protocol was as follows: primary BMECs were inoculated onto 24-well plates until reaching 70-80% confluence. Subsequently, the siRNA was transfected using a Transfection Reagent Kit (ViewSold Biotech, Beijing, China) according to the manufacturer's protocol.

The Cell Proliferation Activity, Cell Cycles, and Apoptosis Assays
The primary BMECs on 24-well plates were exposed to PER2 siRNA and the cell proliferation activity at 12, 24, and 36 h-transfections determined using the Cell Counting Kit-8 (CCK8, Dojindo, Japan) according to the manufacturer's protocols. After 12, 24, and 36 h-transfection, the culture medium was aspirated and the cells were rinsed with PBS and cultured with DMEM-F12, containing a 10% CCK8, for 1.5 h. A 200 µL-culture medium was transferred to 96-well plates to determine absorbance at 450 nm with an MD-SpectraMax M5 plate reader (Molecular Devices Corporation, Sunnyvale, CA, USA).
The cell cycle was assessed with the Cell Cycle and Apoptosis Analysis Kit (C1062, Beyotime, Shanghai, China). The cellular DNA content was quantified via flow cytometry (FACS LSRFortessa, BD, USA) at a 488 nm wavelength and was analyzed with the FlowJo software. The proliferation index (PI) was calculated as: PI = (S + G2/M)/(G0/G1 + S + G2/M) × 100%. Apoptosis was monitored by FITC-conjugated antibodies with Annexin V, using the Annexin V-FITC Apoptosis Detection Kit (C1062, Beyotime, Shanghai, China). Apoptotic cell numbers were determined by FACS LSRFortessa (BD, USA) as described in the manufacturer's protocols. The expression of the apoptosis-related genes caspase-3, caspase-8, and P53 was determined by the real-time PCR and the primers listed in Table S2.

Triacylglycerol Content and Lipid Droplet Determination
The lipid droplet content was determined with red oil O (Sigma, USA). Briefly, transfected cells were fixed in formaldehyde at 4 • C for 1 h and stained with 0.5% oil red O overnight at room temperature. Subsequently, samples were washed with 60% isopropyl alcohol several times. The lipid droplet content was quantified via counting under an inverted microscope (Olympus, Tokyo, Japan, CKX41). To quantify the lipid droplet density, each oil red O optical density (OD) value at 500 nm was measured, after de-staining with isopropanol for 30 min, using a MD-SpectraMax M5 plate reader (Molecular Devices Corporation, CA, USA).
The triacylglycerol (TAG) content in primary BMECs was determined with the Tissue Triglyceride Assay Kit (Applygen, Beijing, China) following the manufacturer's instructions. The specific steps were as follows: cells were lysed and centrifuged at 2000× g for 5 min to harvest the supernatant. The supernatant was incubated at 70 • C for 10 min and was then collected for an enzyme activity measurement. The samples, and a 4 mM glycerol standard at different dilutions, were measured at 550 nm with a microplate reader. The TAG content in samples was calculated using a standard curve, following the manufacturer's protocols.
To investigate lipogenesis at the transcription level, the gene expressions of PPARG, SREBF1, mTOR, ACACA, FAS, LPIN1, SCD, and LPL were determined using the real-time PCR. Primers for these genes are listed in Table S2.

Statistical Analyses
SPSS software version 22.0 was used for statistical analyses. Each experiment was performed in triplicate on different days with three experimental replicates at each time. Differences in cell cycles, cell apoptosis, and lipid synthesis between the PER2 silenced and non-transfected groups were analyzed via a Student's t test. The differences in the target gene expression between PER2 silencing, the negative control, and the non-transfected group were evaluated by a one-way ANOVA. Statistical significance was declared at p < 0.05.

Screening for Transfection Efficiency
The BMECs transfected with siRNA-PER2-b for 36 h had the lowest PER2 mRNA abundance, with an 84.7% inhibitory effect compared to normal BMECs (0 h) ( Figure 1). Thus, all subsequent tests involving PER2 silencing were performed using BMECs transfected with siRNA-PER2-b for 36 h. was then collected for an enzyme activity measurement. The samples, and a 4 m erol standard at different dilutions, were measured at 550 nm with a microplate The TAG content in samples was calculated using a standard curve, following the facturer's protocols.
To investigate lipogenesis at the transcription level, the gene expressions of SREBF1, mTOR, ACACA, FAS, LPIN1, SCD, and LPL were determined using the r PCR. Primers for these genes are listed in Table S2.

Statistical Analyses
SPSS software version 22.0 was used for statistical analyses. Each experim performed in triplicate on different days with three experimental replicates at ea Differences in cell cycles, cell apoptosis, and lipid synthesis between the PER2 s and non-transfected groups were analyzed via a Student's t test. The difference target gene expression between PER2 silencing, the negative control, and the no fected group were evaluated by a one-way ANOVA. Statistical significance was d at p < 0.05.

Screening for Transfection Efficiency
The BMECs transfected with siRNA-PER2-b for 36 h had the lowest PER2 abundance, with an 84.7% inhibitory effect compared to normal BMECs (0 h) (Fi Thus, all subsequent tests involving PER2 silencing were performed using BMEC fected with siRNA-PER2-b for 36 h.

The Cell Proliferation Activity and Cell Cycles
The cell proliferation activity did not differ after the transfection of PER2-siR 12, 24, and 36 h (p > 0.05) (Figure 2A). Based on the cell cycle analysis ( Figure 2B,C silencing increased cell numbers in the G2/M phase (p < 0.05), but there was no sig effect on cells in the G0/G1 and S phases (p > 0.05). The proliferation index (PI) wa

The Cell Proliferation Activity and Cell Cycles
The cell proliferation activity did not differ after the transfection of PER2-siRNA for 12, 24, and 36 h (p > 0.05) (Figure 2A). Based on the cell cycle analysis ( Figure 2B,C), PER2 silencing increased cell numbers in the G2/M phase (p < 0.05), but there was no significant effect on cells in the G0/G1 and S phases (p > 0.05). The proliferation index (PI) was higher in the PER2 silence group (43.501%) than the non-transfected group (34.282%), but there was no significant difference (p > 0.05). in the PER2 silence group (43.501%) than the non-transfected group (34.282%), but there was no significant difference (p > 0.05).

Cell Apoptosis
The cell apoptosis index (the proportion of apoptotic cells among the total counted cells) was not affected by PER2 silencing (p > 0.05) ( Figure 3A). Among the genes measured ( Figure 3B), caspase-8 mRNA expression was upregulated by PER2 silencing (p < 0.05), but its downstream gene, caspase-3, did not differ between the groups (p > 0.05). This indicated that PER2 silencing might have upregulated caspase-8 but failed to drive caspase-3 expression, hence preventing an increase in apoptosis.

Cell Apoptosis
The cell apoptosis index (the proportion of apoptotic cells among the total counted cells) was not affected by PER2 silencing (p > 0.05) ( Figure 3A). Among the genes measured ( Figure 3B), caspase-8 mRNA expression was upregulated by PER2 silencing (p < 0.05), but its downstream gene, caspase-3, did not differ between the groups (p > 0.05). This indicated that PER2 silencing might have upregulated caspase-8 but failed to drive caspase-3 expression, hence preventing an increase in apoptosis.

Lipid Synthesis
The concentrations of triacylglycerol, lipid droplets, and lipogenic gene expressions were measured as indicators of lipid sythesis. Figure 4A depicts lipid droplets in BMECs by oil red O staining. In comparison with the non-transfected group, PER2 silencing led to a 29.8% decrease in lipid droplet density and 38.86% decrease in lipid droplet quantity (p < 0.05) ( Figure 4B,C). Furthermore, the cellular triacylglycerol concentration (p < 0.05) significantly decreased by 48.93% after PER2 silencing ( Figure 4D). At the transcription level, the expression of the lipid transcription regulators PPARG and SREBF1 was downregulated after PER2 silencing in comparison with the non-transfected group (p < 0.05). The expression of the lipid synthesis-related genes ( Figure 5) mTOR, ACACA, FAS, LPIN1, and SCD was downregulated by PER2 silencing (p < 0.05), while LPL mRNA expression was not affected (p > 0.05).

Lipid Synthesis
The concentrations of triacylglycerol, lipid droplets, and lipogenic gene expressions were measured as indicators of lipid sythesis. Figure 4A depicts lipid droplets in BMECs by oil red O staining. In comparison with the non-transfected group, PER2 silencing led to a 29.8% decrease in lipid droplet density and 38.86% decrease in lipid droplet quantity (p < 0.05) ( Figure 4B,C). Furthermore, the cellular triacylglycerol concentration (p < 0.05) significantly decreased by 48.93% after PER2 silencing ( Figure 4D). At the transcription level, the expression of the lipid transcription regulators PPARG and SREBF1 was downregulated after PER2 silencing in comparison with the non-transfected group (p < 0.05). The expression of the lipid synthesis-related genes ( Figure 5) mTOR, ACACA, FAS, LPIN1, and SCD was downregulated by PER2 silencing (p < 0.05), while LPL mRNA expression was not affected (p > 0.05).

Discussion
The milk fat content in dairy cows follows a rhythmic pattern and is thought to be regulated by circadian rhythms. The present study was an investigation to better understand the role of the clock gene PER2 on milk fat synthesis gene networks in BMECs. We employed PER2 silencing experiments in BMECs and observed that PER2 silencing led to a significant inhibition of lipid synthesis-related gene expression. A PER2 silencing model was proposed based on the results from the present study ( Figure 6). The findings suggest that PER2 participates in the coordination of mammary lipid metabolism and may be a component of the control of lipid synthesis in ruminant mammary cells. Thus, these data provide an important reference to molecular links between the circadian clock and lipid metabolism in bovine mammary cells.

PER2 Silencing Downregulates Mammary Lipogenic Genes
PPARG and SREBF1 play a central role in lipid metabolism. In this study, the downregulation of the lipid-related transcription regulators PPARG and SREBF1 in BMECs after PER2 silencing underscored the potential for PER2 to participate in the regulation of mammary lipid synthesis. Consistent with our findings, previous studies revealed that circadian clock genes control lipid metabolism in non-ruminants [20][21][22]. PER2 could directly control PPARG expression as demonstrated by Grimaldi et al. [14] where the knockout of PER2 inhibits lipid metabolism by directly downregulating PPARG in mice. Furthermore, a concerted function between SREBF1 and PPARG was reported in regulating lipid synthesis in ruminants [23]. Thus, by inhibiting PPARG and SREBF1 in BMECs, PER2 could alter various aspects of lipid synthesis in BMECs.
A recent study from our laboratory (not published) aimed to determine circadian rhythms in mRNA expressions of PER2 and lipogenic genes in normal BMECs. We found the expression of both PER2 and PPARG kept decreasing between 12 and 24 h after determination. This suggested the existence of a synchrony of PER2 and PPARG expression in primary BMECs. We further analyzed the rhythms of SREBF1 and observed that the decrease in SREBF1 occurred approximately 4 h after the decrease of PER2 and PPARG, with the decrease continuing between 16 and 24 h after analysis. Together, these findings are consistent with the present study, indicating a similar pattern of change with the circadian rhythm in PER2, SREBF1, and PPARG expression.
was proposed based on the results from the present study ( Figure 6). The findi that PER2 participates in the coordination of mammary lipid metabolism an component of the control of lipid synthesis in ruminant mammary cells. Thus provide an important reference to molecular links between the circadian clo metabolism in bovine mammary cells.

PER2 Silencing Downregulates Mammary Lipogenic Genes
PPARG and SREBF1 play a central role in lipid metabolism. In this study regulation of the lipid-related transcription regulators PPARG and SREBF1 in ter PER2 silencing underscored the potential for PER2 to participate in the r mammary lipid synthesis. Consistent with our findings, previous studies re circadian clock genes control lipid metabolism in non-ruminants [20][21][22]. PER rectly control PPARG expression as demonstrated by Grimaldi et al. [14] wher out of PER2 inhibits lipid metabolism by directly downregulating PPARG in thermore, a concerted function between SREBF1 and PPARG was reported in lipid synthesis in ruminants [23]. Thus, by inhibiting PPARG and SREBF1 in BM could alter various aspects of lipid synthesis in BMECs.
A recent study from our laboratory (not published) aimed to determin rhythms in mRNA expressions of PER2 and lipogenic genes in normal BMEC the expression of both PER2 and PPARG kept decreasing between 12 and 24 h mination. This suggested the existence of a synchrony of PER2 and PPARG e primary BMECs. We further analyzed the rhythms of SREBF1 and observed crease in SREBF1 occurred approximately 4 h after the decrease of PER2 and P the decrease continuing between 16 and 24 h after analysis. Together, these : increase : inhibition Data indicated, at least in non-ruminants, that the expression of SREBF1 and PPARG is partly under the control of the mechanistic target of mTOR [24,25]. Non-ruminant studies revealed that the mTOR pathway could regulate lipid synthesis by integrating various molecular signals culminating in the activation of adipogenesis/lipogenesis through SREBF1 and PPARG [26][27][28][29]. mTOR is essential for regulating SREBF1 at both transcriptional and post-translational levels [30][31][32], and mTORC2 can control lipid synthesis in BMECs by regulating the lipogenic gene expression through PPARG in ruminants [33]. In this study, the downregulation of mTOR at the mRNA level, as well as SREBF1 and PPARG, indicated a strong mechanistic link, whereby PER2 downregulated SREBF1 and PPARG through mTOR.

PER2 Silencing Suppresses Cellular TAG Accumulation and Lipid Droplet Formation
The TAG and lipid droplet levels reflected the lipid synthesis capacity in BMECs. Lipid synthesis included the de novo synthesis of fatty acids and the esterification of newly synthesized fatty acids into TAG [34]. Lipid droplets in BMECs contained a cytosolic TAG-rich core, which is the end result of lipogenesis and esterification. The lower mRNA expression of TAG-related genes (SCD, ACACA, LPIN1 and FAS) and lipid droplet secretionrelated genes (LPIN1) when PER2 was silenced represented a possible mechanism for the decrease in TAG and lipid droplet content. Thus, the fact that PER2 silencing decreased lipid droplets and cellular TAG concentrations in BMECs is further evidence that this clock gene can regulate the basic mechanisms associated with lipid synthesis in BMECs.
The downregulation in the expression of FAS and ACACA, as well as SCD, when PER2 was silenced, indicated that the inhibition of de novo fatty acid synthesis by PER2 silencing contributed to the lower TAG content in BMECs. Both PPARG [35][36][37] and SREBF1 [38][39][40] can regulate TAG synthesis. Our findings were consistent with previous studies where the inhibition of SREBF1 activity was associated with the downregulation Biology 2021, 10, 1226 9 of 11 of SCD, ACACA, and LPIN1 [25,41]. In accordance with the increased activity of PPARG, it led to the upregulation of SCD and LPIN1 expression [35]. LPIN1, a target gene of PPARG in ruminants [23], is one key gene associated with lipid droplet formation. The downregulation of LPIN1, when PER2 was silenced, agreed with the decrease in PPARG expression and the content of lipid droplets, indicating that PER2 also exerted its effect on lipid droplet formation by regulating LPIN1. Collectively, the decrease in lipid droplets and cellular TAG concentrations in BMECs by PER2 silencing was consistent with the mRNA expression data. Thus, this clock protein is associated with various aspects of lipid synthesis in BMECs.

PER2 Silencing Regulates the Cell Cycle, but Not Apoptosis
Cells operate in strict accordance with the G1-S-G2-M cycle and maintain normal biological rhythms under the control of the cell cycle network, Cyclins-CDKs-CKIs [17]. In this study, we observed that the proliferation index was not significantly influenced by PER2 silencing. Furthermore, the increased number of cells in the G2/M phase suggested that PER2 silencing can upregulate the DNA proliferation activity and cellular mitosis to induce more cells into the G2/M. Consistent with our findings, there is compelling evidence for a mechanistic link between PER2 and the cell cycle [42,43], where the downregulation of PER2 accelerated the growth of breast cancer cells [44]. Our results showed that PER2 silencing did not affect cell apoptosis, which was consistent with Hu et al. [19]. Caspase-8, as the initiator of apoptosis at the top of the caspase signaling cascade [45], was significantly upregulated by PER2 silencing, although its downstream gene caspase-3 was not significantly influenced. It is possible that the higher expression of caspase-8 failed to drive the upregulation of downstream genes in the caspase signaling pathway, hence explaining the lack of an effect on cell apoptosis when PER2 was silenced.

Conclusions
The clock protein PER2 regulates genes involved in lipid metabolism, including fatty acid de novo synthesis, fatty acid desaturation, TAG accumulation, and lipid droplet secretion in primary BMECs. The inhibition of PPARG and SREBF1 when PER2 is silenced appears to be an important cause for the reduction in lipid synthesis. Our overall data suggests that PER2 in bovine mammary cells plays a role in governing milk fat synthesis directly, or through the activation of the transcriptional regulators PPARG and SREBF1. This study provides molecular evidence in support of a link between circadian clocks and lipid metabolism in bovines.