Intracellular Ca²⁺ signaling and ORAI calcium release-activated calcium modulator 1 are associated with hepatic lipidosis in dairy cattle

Abstract

Fatty liver is a common metabolic disorder afflicting dairy cows during the periparturient period and is closely associated with endoplasmic reticulum (ER) stress. The onset of ER stress in humans and mice alters hepatic lipid metabolism, but it is unknown if such event contributes to fatty liver in dairy cows soon after parturition. ORAI calcium release-activated calcium modulator 1 (ORAI1) is a key component of the store-operated Ca²⁺ entry mechanism regulating cellular Ca²⁺ balance. The purpose of this study was to investigate the role of ORAI1 on hepatic lipidosis via ER stress in dairy cows. Liver tissue biopsies were collected from Holstein cows diagnosed as healthy (n = 6) or with hepatic lipidosis (n = 6). Protein and mRNA abundance of ER stress-related targets, lipogenic targets, or the transcription regulator SREBP1 and ORAI1 were greater in cows with lipidosis. In vitro, hepatocytes were isolated from four healthy female calves and used for culture with a 1.2 mM mixture of fatty acids (oleic, linoleic, palmitic, stearic, and palmitoleic acid) for various times (0, 3, 6, 9, or 12 h). As incubation time progressed, increases in concentration of Ca²⁺ and abundance of protein kinase RNA-like ER kinase (PERK), inositol-requiring protein 1α (IRE1α), and activating transcription factor-6 (ATF6) protein in response to exogenous fatty acids underscored a mechanistic link among Ca²⁺, fatty acids, and ER stress. In a subsequent study, hepatocytes were transfected with small interfering RNA (siORAI1) or the ORAI1 inhibitor BTP₂ for 48 h or 2 h followed by a challenge with the 1.2 mM mixture of fatty acids for 6 h. Compared with control group, silencing or inhibition of ORAI1 led to decreased abundance of fatty acid synthesis (FASN, SREBP1, and ACACA) and ER stress-related proteins in bovine hepatocytes. Overall, data suggested that NEFA through ORAI1 regulate intracellular Ca²⁺ signaling, induce ER stress, and lead to lipidosis in isolated hepatocytes.

Introduction

In early lactation, mobilization of fat reserves in dairy cows helps compensate for the decrease in dry matter intake (DMI) that leads to a deficiency in energy supply; this process leads to an increase in circulating plasma nonesterified fatty acids (Schultz, 1968; Baird, 1982). Uptake of fatty acids by the liver often leads to lipidosis and ketosis (Emery et al., 1992; a hallmark of negative energy balance [NEB] often causing an increase in veterinary costs, decreased milk production, and prolonged calving intervals; Grummer, 2008). Thus, elucidating mechanisms that are altered during onset of fatty liver could lead to identification of effective preventive strategies.

Studies with nonruminants and dairy cows have established that plasma concentrations of fatty acids are the leading cause of hepatic endoplasmic reticulum (ER) stress (Wang et al., 2006; Gentile et al., 2011; Zhu et al., 2019b). In dairy cows, high concentrations of hydroxybutyrate (BHBA) not only induce ER stress in mammary cells, but also lead to greater milk fat during ketosis (Zhang et al., 2020b). In addition, fatty acids induced ER stress in calf hepatocytes and ER stress in the liver of cows with severe fatty liver (Zhu et al., 2019a). Various signaling cascades associated with normal ER cell processes require high concentrations of Ca²⁺ within the ER lumen. The Ca²⁺ content in the ER is sensed via the store-operated Ca²⁺ entry (SOCE; Immler et al., 2018) and the ORAI calcium release-activated calcium modulator 1 (ORAI1), which upon activation allow for increases of intracellular Ca²⁺ levels to maintain Ca²⁺ balance (Soboloff et al., 2012; Shambharkar et al., 2015).

An unfolded protein response (UPR) within the lumen of the ER, that is, to restore homeostasis, is a typical response to alterations in Ca²⁺ homeostasis or other external stimuli (Fu et al., 2012; Zhu et al., 2016; Yu et al., 2018). Three canonical branches compose the UPR and are characterized by the proteins that exert control: protein kinase RNA-like ER kinase (PERK; also known as eukaryotic translation initiation factor 2 alpha kinase 3, EIF2AK3), inositol-requiring protein 1α (IRE1α; also known as ER to nucleus signaling 1, ERN1), and activating transcription factor-6 (ATF6; Ron and Walter, 2007; Gessner et al., 2014). Specific indicators of ER stress include PERK, IRE1, and ATF6 protein abundance along with mRNA abundance of several downstream genes such as glucose-regulated protein 78 (GRP78; Baiceanu et al., 2016). The C/EBP-homologous protein/growth arrest and DNA damage-inducible gene (CHOP) is an apoptotic transcription factor that is upregulated in response to a cellular stressor.

Studies with nonruminants indicated that ER stress triggers a lipogenic response in the liver due to activation of SREBP-1c and its target genes (Werstuck et al., 2001; Ning et al., 2011). In that context, it is noteworthy that a study in calf primary hepatocytes revealed that high concentrations of fatty acids upregulated de novo lipogenesis including acetyl-CoA carboxylase-α (ACACA) and fatty acid synthase (FASN) abundance thereby contributing to cellular lipid accumulation (Li et al., 2015). The high plasma concentrations of fatty acids in the peripartal period is characterized primarily by palmitic acid (PA), palmitoleic acid (POA), stearic acid (SA), oleic acid (OA), and linoleic acid (LA; Rukkwamsuk et al., 2000; Xu et al., 2016; Liu et al., 2020). Clearly, these encompass the major fatty acid classes, that is, saturated (PA and SA), monounsaturated (POA and OA), and polyunsaturated (LA). Furthermore, these long-chain fatty acids were reported to cause lipid accumulation and injury in primary calf hepatocytes to different degrees (Zhang et al., 2020c). More importantly, silencing of ORAI1 in rat hepatocytes decreased nuclear SREBP abundance and produced a dramatic decrease in lipogenesis (Zhang et al., 2018), suggesting a mechanistic link between Ca²⁺, circulating fatty acid concentrations, and potentially ER stress.

The well-established link between high concentrations of fatty acids and the onset of fatty liver and ketosis led us to hypothesize that there is an association that Ca²⁺ signaling induced ER stress, and then lead to lipidosis in hepatocytes. Thus, the aim of this study was to investigate whether fatty acid-induced ER stress and altered ORAI1 facilitate lipidosis in bovine hepatocytes.

Material and Methods

The Ethics Committee for the Use and Care of Animals, Heilongjiang Bayi Agricultural University (Daqing, China) approved the study protocol (Number of permit: SY201909013).

Animals

Lactating multiparous Holstein cows in this experiment were selected from a 1,000-cow dairy farm with free-stall housing systems located in Daqing, Heilongjiang Province, China. Composition of the lactating cow diets is shown in Table 1. A total of 34 lactating Holstein cows averaging 2.8 ± 0.7 lactations, at 10.13 ± 1.90 days in milk (DIM), and with similar body condition (3.18 ± 0.07, five-point scale) were screened for the study. A preliminary selection of cows was done using blood concentrations of nonesterified fatty acids (healthy concentrations: 0.361 ± 0.128 mmol/L), BHBA (healthy concentrations: BHBA ≤ 1.2 mmol/L), glucose (healthy concentrations: 0.9–5.1 mmol/L), and aspartate aminotransferase (AST; healthy concentrations: 60 ± 9.7 U/L; Table 2). Nonesterified fatty acids, BHBA, glucose, and AST were measured using commercial kits (Beijing Jiuqiang Biotechnology Co. Ltd, Beijing, China) with an automatic clinical analyzer (Synchron DXC800; Beckman Coulter, Inc., Brea, CA). All cows were examined to ensure there were no other clinical disorders. All 34 lactating Holstein cows selected above were then chosen for analysis of hepatic triacylglycerol (TG) content through liver biopsy. Healthy cows were those with hepatic TG content less than 1% (% g/g of wet weight), and cows with fatty liver had hepatic TG content greater than 1% (Bobe et al., 2004). According to a previous study, six cows with fatty liver were chosen based on liver TG content for subsequent experiments (Table 3; Loor et al., 2007) and six healthy cows served as controls.

Table 1.

Composition of lactation diets fed to cows

Item	Lactation
Leymus chinensis	5.50
Whole-plant corn silage	15.0
Corn
Soybean meal	0.700
Lignin (%)	4.00
Non-fibrous carbohydrates (%)	33.1
Soluble sugar (%)	4.90
Starch (%)	16.9
Soluble fiber (%)	8.50
Vitamin D premix	0.0300
Vitamin A premix	0.0100
Vitamin E premix	0.600
DM, % as fed	54.6
NDF, % of DM	39.6
Energy density, NEL/kg DM	1.47
CP, % of DM	13.8

Table 2.

Blood biomarker concentrations in healthy cows and cows with fatty liver

Item	Fatty liver, n = 15		Healthy, n = 19
	Median	Interquartile range	Median	Interquartile range
Nonesterified fatty acids, mM	1.10	0.710, 1.55	0.420	0.190, 0.640
β-hydroxybutyrate, mM	2.06	1.20, 3.37	0.640	0.410, 0.930
Serum glucose, mM	2.38	1.99, 2.54	3.40	2.72, 3.72
AST, U/L	124	79.0, 319	75.8	63.0, 87.0

Table 3.

Production performance and biomarkers of energy balance in healthy cows and cows with fatty liver

Item	Fatty liver, n = 6		Healthy, n = 6		P-value
	Median	Interquartile range	Median	Interquartile range
BW, kg	655	635.5, 675.5	644	628, 663.5	0.0223
Milk production, kg/day	37.2	30, 48	34.8	30, 44	0.467
DMI, kg/d	17.5	14.5, 21.3	18.2	16.2, 22.3	0.59
Milk production/DMI	2.13	1.90, 3.37	1.92	1.83, 1.98	0.0043
Serum glucose, mM	2.23	1.99, 2.54	3.22	2.72, 3.72	0.00066
Nonesterified fatty acids, mM	1.11	0.78, 1.55	0.41	0.19, 0.56	0.00075

Liver tissue samples were biopsied from the 11th to 12th right intercostal space twice from each cow by liver puncture needle. Before liver biopsy, the intercostal space was shaved, sanitized with iodine scrub and 75% alcohol, and anesthetized with a subcutaneous injection of 2% lidocaine HCl (Sigma-Aldrich Co., St. Louis, MO). A scalpel blade was used to make a 3-mm stab incision in the skin. The puncture needle was then thrusted into the liver. The liver tissue (about 200 mg from every collection) was immediately placed in liquid nitrogen until protein and mRNA abundance analyses (Du et al., 2018).

Isolation, culture, and treatment of primary calf hepatocytes

Primary calf hepatocytes were isolated from four healthy female Holstein calves (1-d-old, 40–50 kg, fasting, rectal temperature 38.7–39.7°C) purchased from a commercial dairy farm (Daqing, China) using a modified two-step collagenase IV perfusion method as previously described (Gao et al., 2018). In short, the caudate lobe was obtained from the liver through surgical excision, and hepatocytes isolated under sterile conditions. Blood vessels were intubated and perfused with perfusion solution A (6.7 mM KCl, 140 mM NaCl, 2.5 mM glucose, 10 mM HEPES, and 0.5 mM EDTA, pH 7.4; 37°C, 50 mL/min for 10 min) and solution B (6.7 mM KCl, 140 mM NaCl, 2.5 mM glucose, 30 mM HEPES, and 5 mM CaCl₂, pH 7.4; 37°C, 50 mL/min for 6 min) until the liquid became clear. Subsequently, the liver was perfused with digestion solution (0.1 g of collagenase IV dissolved in 0.5 L of perfusion solution B, pH 7.2–7.4; 20 mL/min for 12 min) to dissociate the hepatic structure. Then, 50 mL of fetal bovine serum (FBS; Hyclone Laboratories, Logan, UT) was added to end the digestion. The liver was cut into small pieces, and the hepatic capsule removed, blood vessels, connective tissue, and incompletely digested tissue. The remaining hepatic parenchyma was filtered sequentially with 50 mesh (300 μm), 100 mesh (150 μm), and 200 mesh (75 μm) cell sieves. The obtained hepatocytes were washed with RPMI-1640 basic medium (Hyclone Laboratories) and centrifuged for 5 min at 500 × g at 4°C. Cell viability was assessed with the Trypan blue dye (Sigma-Aldrich) exclusion method after isolation. Only cells with viability >95% were used for further experiments.

Primary calf hepatocytes were cultured in six-well plates at 2 × 10⁶ cells/mL using RPMI-1640 basic medium supplemented with 10% FBS, 10 μM of insulin, 10⁻⁶ μM of dexamethasone, 10 μg/mL of vitamin C for cell adherence, and incubated at 37°C in 5% CO₂. After 4 h, the adherent medium was replaced with growth medium (RPMI-1640 basic medium supplemented with 10% FBS). The growth medium was replaced with fresh medium every 24 h.

Briefly, after 44 h culture, calf hepatocytes were serum-free starved in RPMI-1640 basic medium for 12 h. Subsequently, hepatocytes were cultured in RPMI-1640 basic medium containing 2% BSA and treated with 1.2 mM fatty acid for another 6 h. Components and concentrations of fatty acids used in this study were chosen based on well-established hematology standards of dairy cows with severe fatty liver (Bertics et al., 1992). A stock fatty acid solution was prepared by diluting fatty acid in 0.1 M KOH at 60°C (pH 7.4). The stock fatty acid (52.7 mM) solution included OA (cis9-18:1, 22.9 mM, 43.5%; Sigma-Aldrich), LA (cis9, cis12-18:2, 2.6 mM, 4.93%; Sigma-Aldrich), PA (16:0, 16.8 mM, 31.9%; Sigma-Aldrich), SA (18:0, 7.6 mM, 14.4%; Sigma-Aldrich), and POA (cis9-16:1, 2.8 mM, 5.31%; Sigma-Aldrich).

Cytosolic calcium

Cytosolic Ca²⁺ concentration was analyzed by flow cytometry. The hepatocyte preparation was washed and stained with 3 μM Fluo-3AM (Biotinium, China) in Tyrode buffer (pH 7.4), cultured at 37°C for 30 min. Relative fluorescence was measured utilizing a Beckman CytoFLEX FCM (Beckman Coulter, Brea, CA).

Inhibition of ORAI1 and transfections

Before treatment with the ORAI1 inhibitor, primary calf hepatocytes were cultured at a density of 1 × 10⁶ cells/cm² in six-well plates with RPMI-1640 basic medium containing 10% FBS, 100 μg/mL streptomycin, and 100 U/mL penicillin at 5% CO₂ with 37°C for 48 h. Hepatocytes were then divided into four groups: control, fatty acid, BTP₂, and BTP₂ + fatty acid. In the BTP₂ and BTP₂ + fatty acid group, 20 μM of N-(4-[3,5-bis(trifluoromethyl)-1H-pyrazol-1-yl]-phenyl)4-methyl-1,2,3-thiadiazole-5-carboxamide (BTP₂; Sigma, Beijing, China) was used for 2 h prior to fatty acid challenge.

For transient transfections, 1 × 10⁶ cells were seeded in six-well plates for 48 h before the experiment. Cells were transfected with siORAI1 using siRNA-mate (Shanghai GenePharma Co., Ltd., Shanghai, China) according to the manufacturer’s protocol. The ORAI1 small interfering (si) RNA was purchased from GenePharma.

Immunofluorescence of ORAI1 protein

Hepatocytes were plated onto six-well plates and used for immunostaining after transfection with vectors. Cells were then washed with PBS and fixed with 4% paraformaldehyde for 30 min at room temperature. Hepatocytes were incubated with 3% Albumin Fraktion V (BioFRoxx, Shanghai, China), 5% normal goat serum (Boster, Wuhan, China), and 0.5% Triton X-100 in PBS (BioFRoxx, Shanghai, China) for 30 min at room temperature to block unspecific binding. Cells were then exposed to ORAI1 rabbit polyclonal antibody (1:500, ProteinTech, China) at 4°C in a humidified chamber overnight. Cells were then rinsed four times with PBS and incubated with Cy3-labeled goat anti-rabbit IgG (1:500, Beyotime Biotechnology, Shanghai, China) for 1 h at room temperature. After four washing steps, the nuclei were stained with Hoechst 33342 (Beyotime Biotechnology) for 8 min at room temperature. After washing three times and covering with a glass coverslip, the image was collected with a laser confocal microscope (LSM 5 PASCAL, Zeiss, Oberkochen, Germany). The fluorescence intensity of ORAI1 was quantified using Image J software (National Institutes for Health, Bethesda, MD).

Lipid droplet fluorescence assay

After hepatocytes were stimulated with 1.2 mM fatty acid at 0, 3, 6, 9, and 12 h, they were washed twice with PBS for 5 min, fixed with 4% paraformaldehyde for 30 min, and then gently washed thrice with PBS. To assay lipid droplet fluorescence, cells were stained with BODIPY 493/503 (Invitrogen). Cells were then washed with PBS for three times in the dark. Nuclei were then stained with Hoechst (Beyotime Biotechnology) for 8 min at room temperature. Fluorescence was determined using an inverted microscope (IX73, Olympus, Tokyo, Japan).

TG content determination

Liver tissue (approximately 20 mg) was mixed with reagent 1 from the enzymatic kit and homogenized in an ice bath. Samples were then vortexed and centrifuged at 8,000 × g for 10 min at 4°C. The TAG content was determined following the kit manufacturer’s instructions (Solarbio, Beijing, China). Total protein concentration was measured using a protein assay kit (Beyotime Biotechnology).

RNA extraction and PCR

Total RNA was isolated from the liver tissue homogenate or hepatocytes using TRIzol (Invitrogen Corporation, China) according to the manufacture’s protocol (https://assets.thermofisher.com/TFS-Assets/LSG/manuals/trizol_reagent.pdf); mRNA was reverse-transcribed with Reverse Transcriptase M-MLV (RNase H-; Takara Bio, Inc., Beijing, China) using an oligodT primer. Calculated mRNA abundance levels in each sample were normalized to β-actin. Abundance was quantified with the method of 2^−ΔΔCT. Reactions were performed in a BioRad iCycler iQTM Real-Time PCR Detection System (Bio-Rad Laboratories Inc., Hercules, CA). The primers used for ORAI1, PERK, IRE1, GRP78, ATF6, CHOP, SREBP1, FASN, ACACA, and TBP are shown in Table 4.

Table 4.

Sequences of primers used for real-time PCR amplification

Gene	Primer, 5′–3′	Primer length	Tm	Gene bank accession number
ORAI1	Forward: TTTGCCGTCCACTTCTAC Reverse: CCTCTTTCCTCCACTTTCT	18 19	54.55 53.63	NM_001099002.1
β-actin	Forward: GCTAACAGTCCGCCTAGAAGCA Reverse: GTCATCACCATCGGCAATGAG	22 21	62.35 59.40	NM_173979.3
GRP78	Forward: GCATCGACCTGGGTACCACCTA Reverse: CCCTTCAGGAGTGAAAGCCACA	22 22	63.51 62.47	NM_001075148.1
ATF6	Forward: AGCCCTGATGGTGCTAACTGA Reverse: TTCATGATTTAACCTGAGAGATTCTGTT	21 28	61.18 59.17	XM_024989877.1
IRE1	Forward: TCCTCCCAGATCCCAACGAT Reverse: ATGCCATCTGAACTTCGGCA	20 20	60.03 60.04	XM_024980955.1
PERK	Forward: GCCGCTCAGCTCTCCTAGTCC Reverse: TGGCTCTCGGATGAACTGGTCTG	21 23	64.30 64.11	NM_001098086.1

Protein extraction and western blotting

Liver (approximately 30 mg) and total protein from primary calf hepatocytes was dissolved using RIPA buffer (Beyotime Biotechnology, Jiangsu, China) include protease inhibitors, put on ice for 30 min, and centrifuged at 14,000 × g with 4°C for 5 min. Protein content was determined using the BCA protein assay kit (Beyotime). Subsequently, equal amounts of protein (30 μg/lane) were separated on 10% SDS-PAGE and electro-transferred onto polyvinylidene difluoride membrane (Millipore corp., Billerica, MA). The membrane was then blocked for 1 h at room temperature in 0.1% Triton-X/PBS containing 5% skim milk powder (blocking buffer). The blocked membranes were incubated overnight at 4°C with specific antibodies for ORAI1 (1:1000, 66223-1-Ig; ProteinTech), PERK (1:1000, C33E10; Cell Signaling, Shanghai), IRE1 (1:1000; ab37073, Abcam, Cambridge, MA), GRP78 (1:250, sc-376768; Santa Cruz, CA), ATF6 (1:1000; ab203119, Abcam, Cambridge, MA), CHOP (1:1000, L63F7; Cell Signaling), SREBP-1c (1:1000, NB100-2215; Novus), FASN (1:1000, C2065; Cell Signaling), and ACACA(1:1000; ab45174, Abcam, Cambridge, MA). Immunoreactive bands were detected using an enhanced chemiluminescence solution (Beyotime). Subsequently, the membranes were incubated with HRP-conjugated secondary antibodies (3:5000; Beyotime) for 30 min at room temperature. The protein abundance signals were visualized by ECL (Beyotime). Lastly, the bands were visualized using a protein simple imager (ProteinSimple, San Jose, CA) and band intensity was quantified using the Image Lab software.

Statistical analysis

The postpartal days and parity among the cows used are expressed as mean ± SD, baseline characteristics of the dairy cows were expressed as the median and interquartile range, and other data were expressed as the means ± SEM, n represents the number of independent experiments. Statistical analysis was conducted using SPSS 26.0 Software (IBM, Chicago, IL) and GraphPad Prism program (Prism 8.0; GraphPad Software, San Diego, CA). All data were tested for normality and homogeneity of variance using the Shapiro–Wilk and Levene tests, respectively. For baseline characteristics of the dairy cows data with skewed distribution, nonparametric statistical analysis was performed using the Wilcoxon test (Zhu et al., 2019a). For other data with Gaussian distribution, parametric statistical analysis was performed using the independent samples t-test for two groups; one-way analysis of variance (ANOVA) was performed for multiple comparisons with Bonferroni correction. Statistical significance was evaluated via unpaired Student’s t-test or one-way ANOVA with a Duncan test for post hoc analysis. Only differences with P-value ≤ 0.05 were considered statistically significant and a P-value ≤ 0.01 was considered highly significant.

Results

Fatty liver and hepatic ORAI1 and ER stress biomarkers

Compared with healthy cows (mean = 0.52% g/g of wet weight), hepatic TG content of cows with fatty liver (mean = 4.12% g/g of wet weight) was greater (P < 0.01). Protein (P < 0.01) and mRNA (P < 0.01) abundance of hepatic SREBP-1c, ACACA, and FASN also were greater in cows with fatty liver (Figure 1A–C). Furthermore, protein (P < 0.05) and mRNA (P < 0.01) abundance of ORAI1 (Figure 1A–C) and mRNA abundance of the ER stress-related genes GRP78, PERK, IRE1, and ATF6 (P < 0.05) were greater in cows with fatty liver (Figure 1A). Similarly, protein abundance of the ER stress-related proteins GRP78, PERK, IRE1, and ATF6 (P < 0.05) was greater in the liver of dairy cows with fatty liver (Figure 1B and C).

Figure 1.

Abundance of FASN, ACACA, PERK, IRE1, GRP78, ATF6, SREBP1, and ORAI1 was increased in dairy cows with fatty liver. (A) Relative mRNA abundance of FASN, ACACA, PERK, IRE1, GRP78, ATF6, SREBP1, and ORAI1. (B) Western blot analysis of FASN, ACACA, PERK, IRE1, GRP78, ATF6, SREBP1, and ORAI1. (C) Relative protein abundance of FASN, ACACA, PERK, IRE1, GRP78, ATF6, SREBP1, and ORAI1. The data of the control group were used to normalize the data of the cows with fatty liver group. The data were analyzed with paired t-tests and expressed as the mean ± SEM (n = 6 per group). *P < 0.05, ^**P < 0.01.

Exogenous fatty acids affect ORAI1 abundance, ER stress induction, cytosolic Ca²⁺, and lipid accumulation

Following treatment with fatty acids, the levels of cytosolic Ca²⁺ concentration increased gradually, reaching a peak at 1 h post-fatty acid treatment (Figure 2). Furthermore, protein abundance of PERK, IRE1, ATF6, and CHOP increased gradually in a time-dependent manner, reaching a peak at 6 h post-fatty acid treatment (Figure 4A and B). Protein abundance and immunofluorescence of ORAI1 both had a similar trend (Figure 3 and 4A and B). Lipid droplet fluorescence staining indicated a peak in lipid droplet formation at 12 h post-fatty acid treatment (Figure 5).

Figure 2.

Effect of fatty acid load on intracellular Ca²⁺ concentration in calf primary hepatocytes. Cells were treated with a mixture of 1.2 mM fatty acid (included cis9-18:1 [0.522 mM], cis9,cis12-18:2 [0.059 mM], 16:0 [0.383 mM], 18:0 [0.173 mM], and cis9-16:1 [0.064 mM]) for 0, 0.5, 1, 3, 6, 9, and 12 h. (A) Cytosolic Ca²⁺ concentration was analyzed by flow cytometry. (B) Quantification of intracellular Ca²⁺ concentration. Comparisons among groups were calculated using a one-way ANOVA with a Duncan correction. The data presented are the mean ± SEM; *P ≤ 0.05, ^**P ≤ 0.01.

Figure 4.

Effect of fatty acid load on ER stress and ORAI1 abundance in calf primary hepatocytes. Cells were treated with 1.2 mM fatty acid for 0, 3, 6, 9, and 12 h. (A) Western blot analysis of PERK, IRE1, GRP78, ATF6, CHOP, and ORAI1. (B) Relative protein abundance of PERK, IRE1, GRP78, ATF6, CHOP, and ORAI1. Comparisons among groups were calculated using a one-way ANOVA with a Duncan correction. The data presented are the mean ± SEM; *P ≤ 0.05, ^**P ≤ 0.01.

Figure 3.

Effect of fatty acid load on immunofluorescence of ORAI1 protein in calf primary hepatocytes. Cells were treated with 1.2 mM fatty acid for 0, 0.5, 1, 3, 6, 9, and 12 h. Immunofluorescence of ORAI1 protein, scale bar = 25 μm.

Figure 5.

The effect of fatty acid on lipid accumulation in calf primary hepatocytes. Cells were treated with 1.2 mM fatty acid for 0, 3, 6, 9, and 12 h. Lipid droplet staining, scale bar = 25 μm. Data of the 0 mM fatty acid group were used to normalize other time points.

ER stress and lipogenic proteins are altered in response to inhibition of ORAI1 in hepatocytes

Both mRNA and protein abundance of ORAI1 in hepatocytes decreased in response to inhibition of ORAI1 (Figure 6A–C; P < 0.01). In addition, inhibition of ORAI1 markedly alleviated the increase in mRNA and protein abundance of UPR indicators including PERK, IRE1, GRP78, and ATF6 induced by fatty acids (Figure 6A–C; P < 0.01). The inhibition also markedly alleviated the increase in protein abundance of FASN, ACACA, and SREBP1 induced by fatty acids (Figure 7; P < 0.01).

Figure 6.

The effect of an ORAI1 inhibitor on ER stress in calf primary hepatocytes. Treatments were control group (cells cultured in RPMI-1640 basic medium for 6 h), 1.2 mM fatty acid (cells treated with 1.2 mM fatty acid for 6 h), BTP₂ (cells treated with 20 μM BTP₂ for 2 h), and BTP₂ + 1.2 mM fatty acid (BTP2 for 2 h prior to fatty acid and then treated with 1.2 mM fatty acid for another 6 h). (A) Relative mRNA abundance of PERK, IRE1, GRP78, ATF6, CHOP, and ORAI1. (B) Western blot analysis of PERK, IRE1, GRP78, ATF6, CHOP, and ORAI1. (C) Relative protein abundance of PERK, IRE1, GRP78, ATF6, CHOP, and ORAI1. The data of the control were used to normalize other treatments. Comparisons among groups were calculated using a one-way ANOVA with a Duncan correction. The data presented are the mean ± SEM; *P ≤ 0.05, ^**P ≤ 0.01 indicate differences from control. ^#P ≤ 0.05, ^##P ≤ 0.01 indicate differences from fatty acid alone.

Figure 7.

The effect of an ORAI1 inhibitor on fatty acid synthesis in calf primary hepatocytes. Treatments were control group (cells cultured in RPMI-1640 basic medium for 6 h), 1.2 mM fatty acid (cells treated with 1.2 mM fatty acid for 6 h), BTP₂ (cells treated with 20 μM BTP₂ for 2 h), and BTP₂ + 1.2 mM fatty acid (BTP2 for 2 h prior to fatty acid and then treated with 1.2 mM fatty acid for another 6 h). (A) Western blot analysis of FASN, ACACA, and SREBP1. (B) Relative protein abundance of FASN, ACACA, and SREBP1. The data of the control group were used to normalize other treatments. Comparisons among groups were calculated using a one-way ANOVA with subsequent Duncan correction. The data presented are the mean ± SEM; *P ≤ 0.05, ^**P ≤ 0.01 indicate differences from control. ^#P ≤ 0.05, ^##P ≤ 0.01 indicate differences from fatty acid alone.

Silencing ORAI1 alters ER stress and lipogenic protein abundance in hepatocytes

Silencing of ORAI1 via transient transfection decreased mRNA and protein abundance of ORAI1 (Figure 8A–C; P < 0.01) in hepatocytes. In addition, it markedly alleviated the increase in mRNA and protein abundance induced by fatty acids of UPR indicators including PERK, IRE1, GRP78, and ATF6 (Figure 8A–C; P < 0.01). Silencing of ORAI1 markedly alleviated the increase in protein abundance of FASN, ACACA, and SREBP1 induced by fatty acids (Figure 9; P < 0.01).

Figure 8.

Effect of ORAI1 silencing on ER stress in calf primary hepatocytes. Treatments were control siRNA group (cells infected with siRNA-mate for 24 h), control siRNA + 1.2 mM fatty acid (cells infected with siRNA-mate for 24 h then treated with 1.2 mM fatty acid for another 6 h), ORAI transient transfection (siORAI1; transient transfection with siORAI1 for 24 h), and siORAI1 + 1.2 mM fatty acid (transfection with siORAI1 for 24 h then treated with 1.2 mM fatty acid for another 6 h). (A) Relative mRNA abundance of PERK, IRE1, GRP78, ATF6, CHOP, and ORAI1. (B) Western blot analysis of PERK, IRE1, GRP78, ATF6, CHOP, and ORAI1. (C) Relative protein abundance of PERK, IRE1, GRP78, ATF6, CHOP, and ORAI1. The data of the control siRNA group were used to normalize other treatments. Comparisons among groups were calculated using a one-way ANOVA with a Duncan correction. The data presented are the mean ± SEM; *P ≤ 0.05, ^**P ≤ 0.01 indicate difference from control. ^#P ≤ 0.05, ^##P ≤ 0.01 indicate difference from siORAI1 + fatty acid alone.

Figure 9.

Effect of ORAI1 silencing on fatty acid synthesis in calf primary hepatocytes. Treatments were control siRNA group, control siRNA + 1.2 mM fatty acid group, siORAI1 group and siORAI1 + 1.2 mM fatty acid group. (A) Western blot analysis of FASN, ACACA, and SREBP1. (B) Relative protein abundance of FASN, ACACA, and SREBP1. The data of the control siRNA group were used to normalize other treatments. Comparisons among groups were calculated using a one-way ANOVA with a Duncan correction. The data presented are the mean ± SEM; *P ≤ 0.05, ^**P ≤ 0.01 indicate differences from control. ^#P ≤ 0.05, ^##P ≤ 0.01 indicate differences from siORAI1 + fatty acid alone.

Discussion

At least in nonruminants, high concentrations of circulating fatty acids directly induce ER stress in hepatocytes (Pagliassotti et al., 2007; Zhang et al., 2018). Various aspects of lipid metabolism rely on proper ER function, where many of the enzymes involved in intermediary and complex lipid metabolism reside, hence, rendering this organelle essential in the control of lipid homeostasis in tissues like the liver (Fu et al., 2012). Assessment of the abundance of ER proteins in the liver of dairy cows during early lactation indicated the existence of ER stress, a response that was associated with greater concentrations of plasma fatty acids (Gessner et al., 2014; Khan et al., 2015). Although data from nonruminants indicated that ER stress is caused by multiple stimuli, a link between plasma fatty acids and ER stress during early lactation in dairy cows cannot be discounted. Thus, we used primary hepatocytes from calves (which are easily isolated and cultured) to perform a challenge with exogenous fatty acids to assess ER stress over time under controlled conditions. Clearly, the time-dependent upregulation of PERK, IRE1, ATF6, and CHOP indicated that fatty acid supply to liver cells can induce hepatic ER stress in dairy cows.

Evidence from nonruminant studies underscored that induction of ER stress precipitates changes in hepatic lipogenesis or lipid accumulation (Baiceanu et al., 2016). For instance, the inhibition of ER stress ameliorated fatty acid-induced lipid accumulation in Huh-7 cells (Kammoun et al., 2009). In the present study, the time-dependent induction of lipid accumulation in calf hepatocytes in response to exogenous fatty acids was similar to a previous study (Kammoun et al., 2009). However, to our knowledge, a potential mechanistic link between fatty acid influx into the liver, ER stress, and lipid accumulation in bovine has not been assessed. Although the fatty acid mixture used to challenge hepatocytes in the present study is similar to published studies with ketotic (Yamdagni and Schultz, 1970) or healthy dairy cows (Li et al., 2012, 2013; Gao et al., 2018; Zhang et al., 2020a), it is possible that varying the profiles of these fatty acids could lead to different results. Thus, care should be taken when extrapolating results from the present study.

The protein ORAI is a plasma membrane Ca²⁺ influx protein belonging to the SOC family. At least in nonruminants, the SOC control the channel-mediated Ca²⁺ influx and signaling and play an important role in multiple physiologic processes (Feske, 2010). The ER acts as a reservoir for Ca²⁺ to help maintain intracellular homeostasis. Upon Ca²⁺ depletion in the ER, ORAI1 channels are activated and promote SOCE (Brandman et al., 2007). In β-TC3 cells, free FA-induced ER stress was mediated by SOCE through ORAI1 (Cui et al., 2013). Those responses were confirmed in the present study and extended to demonstrate a role for ORAI1 on the ER stress response. Because silencing of ORAI1 prevented the increase in abundance of the ER stress marker GRP78, ER the stress sensors PERK, IRE1, ATF6, CHOP, and abundance of lipogenic proteins induced by fatty acids, we speculate that availability of fatty acids within hepatocytes and the subsequent induction of ER stress might be a signal regulating ORAI1. Although we could not discern the exact mechanisms associated with the temporal change in concentration of Ca²⁺ after challenge with exogenous fatty acids, the data suggested that at least during fatty acid stimulation ORAI1 may contribute to ER stress and lipid accumulation.

In addition to the well-established role in Ca²⁺ homeostasis, work with nonruminants indicated that ORAI1 regulates the expression of SREBP1 and FASN in the liver (Zhang et al., 2018). Thus, the upregulation of abundance of lipogenic genes (ACACA and FASN) and the transcription regulator SREBP1 (Li et al., 2015) in the present study confirmed those effects reported previously. To further strengthen the link between ORAI1, ER stress, and lipogenesis, we measured the abundance of SREBP1, ACACA, and FASN in hepatocytes in which ORAI1 was silenced or inhibited. The fact that the upregulation of SREBP1 and its target proteins (ACACA and FASN) induced by exogenous fatty acid challenge was attenuated by silencing or inhibiting ORAI1 led us to speculate that this transporter mediates in part the development of fatty acid-induced ER stress and lipid accumulation in the liver.

Based on the data generated, we proposed an empirical model illustrating molecular mechanism (Figure 10). Briefly, fatty acids activate ORAI1, ER stress, and upregulate the abundance of lipogenic genes and the transcription regulator SREBP1, which lead to lipid accumulation. Therefore, under conditions of fatty acid stimulation, decreased abundance of ORAI1 via silencing or pharmacological inhibition could reduce ER stress and the abundance of ACACA, FASN, and the transcription regulator SREBP1, resulting in a reduction in lipid accumulation.

Figure 10.

Schematic model illustrating effects of fatty acids on ORAI1 and ER stress, abundance of lipogenic genes and the transcription regulator SREBP1. Fatty acids activates ORAI1, ER stress (PERK, IRE1, ATF6, and GRP78) and upregulates the abundance of lipogenic genes (ACACA and FASN) and the transcription regulator SREBP1, which lead to lipid accumulation. Therefore, under conditions of fatty acid stimulation, a decrease in abundance of ORAI1 via silencing or pharmaceutical inhibition could reduce ER stress and the abundance of ACACA, FASN and the transcription regulator SREBP1, resulting in a reduction of lipid accumulation.

Although several published studies addressing mechanistic aspects of adult ruminant liver metabolism have relied on calf hepatocytes (Mashek and Grummer, 2003, Chandler and White, 2017; Zhou et al., 2018; Zhang et al., 2020a), it is known that aspects of hepatic lipid metabolism differ in preruminant vs. mature ruminant animals (Hahn et al., 2017). Thus, caution should be taken when attempting to extrapolate findings using hepatocytes from 1-d-old calves to whole-animal liver metabolism in periparturient cows. Although in vitro data suggest that ORAI1 signaling could affect lipid metabolism and ER function, the relevance of these data to the transition cow remains to be determined.

In conclusion, the present study underscored the existence of a mechanism regulating lipidosis in bovine liver by providing evidence that exogenous fatty acid supply promotes ER stress followed by induction of lipid accumulation in primary hepatocytes. The role of ORAI1 calcium release-activated calcium modulator 1 in this process appears to be through its effect on intracellular Ca²⁺ homeostasis.

Glossary

Abbreviations

ACACA

acetyl-CoA carboxylase alpha

ATF6

activating transcription factor-6

DIM

days in milk

DMI

dry matter intake

ER

endoplasmic reticulum stress

FASN

fatty acid synthase

IRE1α

inositol-requiring protein 1α

NEB

negative energy balance

ORAI1

ORAI calcium release-activated calcium modulator 1

PERK

protein kinase RNA-like ER kinase

SOCE

store-operated Ca2+ entry

SREBP1

sterol regulatory element binding protein 1

TG

triacylglycerol

UPR

unfolded protein response

Author Contributions

B.B.Z., W.Y., J.J.L., M.L., and C.X. designed this project. M.L., Y.Y.Z., A.A., S.W., J.N.W., J.J.W., M.Y.L., Y.F.S., and L.Y.Y. performed the experiments and analyzed the results. J.J.L. revised the manuscript. M.L. and B.B.Z. drafted this paper. All authors have read and approved the final manuscript.

Conflict of interest statement

The authors have no conflicts of interest to declare.