Lubabegron fumarate acts as a β-adrenergic receptor antagonist in cultured bovine intramuscular and subcutaneous adipocytes

Abstract

We hypothesized that lubabegron fumarate (LUB) (Experior, Elanco Animal Health, Greenfield, IN) would act as an antagonist to β-adrenergic receptor (β-AR) subtypes in primary bovine subcutaneous (s.c.) and intramuscular (i.m.) adipocytes differentiated in culture. This study employed LUB, dobutamine (DOB, a selective β₁-agonist), salbutamol (SAL, a selective β₂-agonist), and propranolol (PRO, a non-selective β-AR antagonist). Preadipocytes were isolated by standard techniques from bovine longissimus muscle and overlying s.c. adipose tissue and differentiated to adipocytes for 14 d. The adipocyte source x stage of differentiation interaction was significant for β-adrenergic receptors-1 (ADRB1) (P = 0.001) and ADRB2 (P = 0.01) in that expression of ADRB1 and ADRB2 was greater in s.c. adipocytes than in s.c. preadipocytes; expression of the ADRB1-3 did not change after differentiation of i.m. adipocytes. CCATT/enhancer-binding protein alpha (CEBPA) expression increased upon differentiation in both s.c. and i.m. adipocytes (P = 0.006). The source x stage of differentiation interaction was significant for peroxisome proliferator-activated receptor gamma (PPARG) (P ≤ 0.001) and fatty acid binding protein-4 (FABP4) (P = 0.004). Expression of PPARG increased after differentiation of s.c. preadipocytes to adipocytes, but PPARG expression did not change with differentiation of i.m. preadipocytes to adipocytes. FABP4 expression increased after differentiation of both s.c. and i.m. adipocytes, but FABP4 expression increased to a greater extent in s.c. adipocytes. In s.c. adipocytes, DOB elevated cAMP and glycerol production and protein kinase A (PKA) activity, and SAL increased PKA activity; these effects were abolished by LUB and PRO (P < 0.001). Incubation of i.m. adipocytes with SAL increased cAMP production and PKA activity, which was attenuated by LUB and PRO (P ≤ 0.006). In s.c. adipocytes, SAL, LUB + SAL, and LUB + DOB upregulated hormone sensitive lipase (HSL) (P < 0.001) and perilipin (P = 0.002) gene expression. In i.m. adipocytes, DOB and LUB + DOB increased HSL gene expression (P = 0.001) and LUB + SAL depressed adipose triglyceride lipase expression below control levels (P = 0.001). These results demonstrate that LUB is a β-AR antagonist at the β₁-AR and β₂-AR subtypes in s.c. adipocytes, and that s.c. and i.m. exhibit different responses to β-AA and LUB.

Lay Summary

We hypothesized that lubabegron fumarate (Experior, Elanco, Greenfield, IN) would act as an antagonist to β-adrenergic receptor subtypes in primary bovine backfat (subcutaneous) and marbling (intramuscular) adipocytes differentiated in culture. Fat cells were isolated from marbling of longissimus muscle and overlying backfat. In backfat cells, lubabegron fumarate downregulated genes associated with turnover of stored lipid, and lubabegron fumarate reversed the increase in cyclic AMP and protein kinase A caused by the β₁-adrenergic receptor agonist, dobutamine, and the β₂-adrenergic agonist, salbutamol. Increasing cyclic AMP amount and protein kinase A activity would lead to a decrease in backfat lipid stores (reducing backfat thickness), and this would be effectively blocked by lubabegron fumarate. Salbutamol but not dobutamine increased cyclic AMP amount and protein kinase A activity in marbling fat cells, and this effect was blocked by lubabegron fumarate. Taken together, the results of this study indicate that lubabegron fumarate antagonizes the effects of hormones that promote lipid loss from backfat and marbling. However, marbling fat cells are not as responsive as backfat fat cells to β-adrenergic agonists, so β-adrenergic agonists such as Zilmax and OptiFlex should have less effect on marbling scores than on backfat thickness.

Intramuscular and subcutaneous adipocytes respond differently to β-adrenergic agonists and lubabegron fumarate (Experior, Elanco, Greenfield, IN), so these compounds should have less effect on marbling scores than on backfat thickness.

Introduction

Three β-adrenergic receptors (β-AR) subtypes have been identified in mammalian tissues, β₁-AR, β₂-AR, and β₃-AR (Strosberg, 1997), which either stimulate or inhibit the physiological responses of β-agonists (β-AA) and β-AR antagonists (Woodcock, 2007; Lynch and Ryall, 2008). The β-AR are members of a large superfamily of G protein-coupled receptors (reviewed for livestock species by Johnson et al., 2014). The stimulation of β-AR by β-AA activates the adenylate cyclase-cAMP-protein kinase A (PKA) signaling cascade (Liggett et al., 1993; Mersmann, 1998; Wallukat, 2002). The predominant subtype in bovine adipose tissue is the β₂-AR, whereas β₁-AR is the primary subtype in porcine adipose tissue (Sillence and Matthews, 1994; McNeel and Mersmann, 1999; Hwang et al., 2021).

Treating beef cattle and sheep with β-AA promotes muscle hypertrophy and decreases adipose tissue mass, the latter effect elicited by stimulation of triacylglycerol (TAG) hydrolysis or inhibition of fatty acid synthesis (Emery et al., 1984; Jones et al., 1985; Coleman et al., 1988; Smith et al, 1989; Moloney et al., 1990; Schiavetta et al., 1990; Smith et al., 1995). At present, the commercially available β-AR agonists are zilpaterol hydrochloride (ZIL) and ractopamine hydrochloride (RAC), but recently a new synthetic β-AR ligand, lubabegron fumarate (LUB, Experior, Elanco Animal Health, Greenfield, IN) has been introduced. Using an adipose tissue ex vivo explant in vitro system, we demonstrated that isoproterenol (ISO), dobutamine hydrochloride (DOB, a selective β₁-agonist), and salbutamol sulfate (SAL, a selective β₂-agonist) increased cyclic adenosine monophosphate (cAMP) production in subcutaneous (s.c.) adipose tissue explants but not in intramuscular (i.m.) adipose tissue explants (Hwang et al., 2021), suggesting that i.m. adipose tissue is relatively refractory to β-AA. The increased cAMP production in s.c. adipose tissue explants elicited by ISO, DOB, and SAL was attenuated by LUB, indicating that LUB is a β-AR antagonist. For the current study, we hypothesized that LUB would be a β₁-AR and β₂-AR antagonist in differentiated primary bovine s.c. and i.m. adipocytes in culture. Additionally, we hypothesized that LUB and other β-AA would up- or downregulate the expression of genes associated with β-AR action and hydrolysis of TAG.

Materials and Methods

Institutional approval

All experimental procedures were approved by the Texas A&M University Institutional Animal Care and Use Committee, AUP #2018-036A.

Materials

Lubabegon fumarate was generously provided by Elanco Animal Health (Greenfield, IN). Other drugs and reagents were purchased from the following companies: DOB, SAL, propranolol hydrochloride (PRO), insulin, dexamethasone (DMSO), gentamicin, Glycerol Assay Kit, DNAseI, bovine serum albumin (BSA), fetal bovine serum (FBS), phenymethylsulfony fluoride (PMSF), and phosphatase inhibitor cocktail (Sigma-Aldrich, St. Louis, MO); troglitazone, rosiglitazone, ciglitazone, 3-isobutyl-1-methylxanthine (IBMX), and forskolin (Cayman Chemical, Ann Arbor, MI); Cell Counting kit-8 (Dojindo, Rockville, MD); Cyclic AMP XP Assay Kit (Cell Signaling Technology, Danvers, MA); antibiotic-antimycotic, Dulbecco’s modified Eagle’s Medium, nutrient mixture Ham’s F12 (DMEM/F12), Dulbecco’s phosphate-buffer saline (DPBS), collagenase II, gentamicin, amphotericin B, Protein Kinase A Colorimetric Activity Kit, Pierce BCA Protein Assay Kit, and protease inhibitor (Thermo Fisher Scientific, Waltham, MA); qScript cDNA synthesis kit and Perfecta SYBR Green Fastmix (Quanta Biosciences, Gaithersburg, MD); and DNAseI and HiBind RNA mini column (Omega Bio-tek, Inc., Norcross, GA).

Bovine adipose tissue sampling at slaughter

Angus crossed bred steers were fed a standard, corn-based finishing diet at the Texas A&M University McGregor Research Center, McGregor, TX, and then transported approximately 190 km to the Texas A&M Rosenthal Meat Science and Technology Center, College Station, TX. Cattle were stunned using a captive bolt followed by exsanguination. The longissimus muscle (LM) and overlying s.c. adipose tissue from the 8th to 10th thoracic region of three steers were removed by cutting through the hide immediately lateral to the spinal process. The LM was transported to the laboratory in pre-warmed DPBS containing 3% antibiotic-antimycotic within 30 min post-stunning. LM samples were obtained from grain-finished steers from another investigator’s study and final BW and carcass quality were not recorded.

Primary preadipocyte isolation from bovine adipose tissues

Fresh LM and overlying s.c. adipose tissue were rinsed with pre-warmed DPBS containing 1% antibiotic-antimycotic to remove blood clots. Subcutaneous and i.m. adipose tissues were dissected fresh from the LM muscle as described previously (Hwang et al., 2021). Adipose tissues were minced into small pieces using surgical scissors and incubated in pre-warmed digestion buffer containing DMEM/F12 (1:1 vol/vol), 0.15% collagenase II, and 1% BSA (essentially fatty acid free) for 45 min at 37 °C in a shaking incubator. Following incubation, the digest was filtered through a 100-μm nylon mesh to eliminate undigested adipose tissues, mature adipocytes, and large cell aggregates. The filtered stromal vascular fractions were centrifuged at 1,500 rpm for 10 min at room temperature and then were incubated in erythrocyte lysis buffer for additional 5 min at room temperature to lyse red blood cells. The suspension was centrifuged at 1,500 rpm for 10 min and resuspended in fresh growth medium (DMEM/F12, 10% FBS, and 1% antibiotic-antimycotic). Preadipocytes were counted and directly cultured at 37 °C in a humidified incubator with 5% CO₂ or were frozen in liquid nitrogen until further use.

Preadipocyte differentiation

Subcutaneous and i.m. preadipocytes were cultured in normal growth media until 80% confluent. Cells were differentiated for 14 d into mature adipocytes in DMEM/F12 containing 2% FBS, 1 μM DMSO, 20 μg/mL insulin, 0.5 mM IBMX, 50 µg/mL gentamicin, 2.5 µg/mL amphotericin B, and 1% antibiotic-antimycotic supplemented with peroxisome proliferator-activated receptor gamma (PPARγ) agonists (10 μM troglitazone, 10 μM ciglitazone, and 1 μM rosiglitazone) for the first 8 d. From days 8 to 14, maintenance medium containing 20 μg/mL insulin and 1 μM rosiglitazone was used. Media were replaced every 48 h. Morphologic changes of the s.c. and i.m. adipocytes were monitored under a light microscope.

Cell viability (mitochondrial dehydrogenase) assay

Subcutaneous preadipocytes were seeded at a density of 5 × 10³ cells/well in 96-well microplates and differentiated for 12 d as described above. Adipocytes were incubated with 0 or 10⁻⁹ to 10⁻⁴ M DOB, SAL, or LUB for 1 h. Viabilities were evaluated by the CCK-8 cell counting kit according to the manufacturer’s instructions. Eight wells were used for each group of incubations. Following treatment, 10 µL of CCK-8 was added into each well, and then the cells were incubated for additional 2 h in an incubator at 37 °C. The absorbance of the formazan product was measured at 450 nm using an Epoch microplate reader (Biotek Instruments, Winooski, VT). The CCK-8 assay measures mitochondrial dehydrogenase (MDH) activity, which is taken as a measure of the number of viable cells (https://www.dojindo.eu.com/store/p/456-Cell-Counting-Kit-8.aspx. The MDH activity assay may not truly indicate number of viable adipocytes, which is discussed below.

cAMP assay

Production of cAMP was determined by using the Cyclic AMP XP Assay Kit according to the manufacturer’s instructions. Briefly, s.c. and i.m. preadipocytes were seeded at a density of 2 × 10⁴ cells/well in 24-well microplates. At day 14 of differentiation, s.c. and i.m. adipocytes were starved for 1 h in DMEM/F12 containing 0.5% BSA, 0.5 mM IBMX, 10 μM forskolin, and 1% antibiotic-antimycotic at 37 °C. After starvation, adipocytes were pre-incubated in the presence or absence of either 1 μM LUB or 10 μM PRO for 5 min. Following 5 min, the cells were incubated with 10 μM DOB, 10 μM SAL, or 1 μM LUB for 15 min. Adipocytes were lysed in 200 μL cell lysis buffer provided by the supplier in addition to 1 mM PMSF and 10 μL/mL protease inhibitor. Fifty microliters of standards or cell lysates from s.c. and i.m. adipocytes were co-incubated in 96-well plates with the horseradish peroxidase-linked cAMP substrate coated onto an immobilized rabbit monoclonal cAMP antibody at room temperature for 3 h on a horizontal orbital plate shaker. After the reaction, color development was measured at 450 nm using an Epoch microplate reader. A standard curve was generated by a four-parameter logistic regression, and cAMP concentrations were calculated using GraphPad Prism 6.04 software (GraphPad Software Inc., San Diego, CA). Protein concentrations were determined by Pierce BCA Protein Assay Kit. Concentrations of cAMP were normalized to cellular protein content and values are reported as pmol cAMP/mg protein.

Glycerol assay

Glycerol was measured in media from the s.c. and i.m adipocyte cultures after the 14-d differentiation period, using a glycerol assay kit according to the manufacturer’s procedures. In brief, 10-µL media were reacted with 100-μL glycerol reaction reagent for 20 min at room temperature and the absorbance was measured at 570 nm. The glycerol standard was used to generate a calibration curve. Glycerol release was normalized to cellular protein content and values are reported as pmol glycerol/mg protein.

Protein kinase A activity

Protein kinase A (PKA) activity was measured in a different set of differentiated adipocytes using PKA Colorimetric Activity Kit following manufacturer’s protocol. Briefly, differentiated adipocytes in 24-well plates were lysed in 200 μL of activated cell lysis buffer provided by the supplier in addition to 1 mM PMSF, 1 μL/mL protease inhibitor, and 1 μL/mL phosphatase inhibitor cocktail. Supernatant fractions were diluted 1:1 in kinase assay buffer provided by the supplier, to which 1 mM PMSF and 0.5 μL/mL protease inhibitor were added. Forty microliters of standards samples and 10 μL ATP were co-incubated in 96-well plates at 30 °C for 90 min on a horizontal orbital plate shaker. Twenty-five microliters of rabbit phospho-PKA substrate antibody and 25 μL of secondary goat anti-rabbit IgG-HRP (supplied by the manufacturer) were co-incubated at room temperature for 60 min with shaking. Color development was measured at 450 nm using an Epoch microplate reader. A standard curve was generated by a four-parameter logistic regression, and the PKA activity was calculated using GraphPad Prism 6.04 software.

Quantitative real-time polymerase chain reaction

Gene expression was measured by quantitative real-time polymerase chain reaction (RT–qPCR). Subcutaneous and i.m. preadipocytes were seeded at a density of 5 × 10⁵ cells in T-25 flasks. At day 14 of differentiation, adipocytes were incubated with 10 μM DOB, 10 μM SAL, 1 μM LUB, 1 µM LUB + 10 µM DOB, or 1 µM LUB + 10 µM SAL for 3 d in DMEM/F12 containing 2% FBS and 1% antibiotic-antimycotic at 37 °C. Total RNA was isolated from differentiated s.c. and i.m. adipocytes using HiBind RNA mini columns followed by DNAseI treatment. Total RNA was quantified on the NanoDrop 2000 Spectrophotometer (Thermo Fisher Scientific, Waltham, MA) and reverse transcription was performed with the qScript cDNA synthesis kit. Expression of genes related to β-AR activity and adipocyte differentiation were analyzed in a CFX384^TM Real-Time System (Bio-rad, Hercules, CA) using the Perfecta SYBR Green fastmix kit. Different efficacies of cDNA synthesis between samples were normalized with three reference genes, ribosomal protein 9 (RSP9), glyceraldehyde-3-phosphate dehydrogenase (GAPDH), and succinate dehydrogenase (SDHA). Relative mRNA amount was determined by the cycle threshold (CT) deviation of an unknown sample vs. geometric mean of three reference genes and data are presented as either 2^−ΔCT (β-AR) or 2^−ΔΔCT (other genes). Primers are listed in Table 1.

Table 1.

Primers for RT–qPCR

Gene1	Accession number	Sequence	Amplicon length (bp)
ABHD5	NM_001076063.2	F: 5ʹ-CATCCAGGGTTAGTCATCTCATT-3ʹ R: 5ʹ-CCAAGGCTCTGATCCAAACT-3ʹ	105
ADRB1	NM_194266.1	F: 5ʹ-AGAAGGCACTCAAGACGCT-3ʹ R: 5ʹ-CACCACGTTGGCTAGGAAGA-3ʹ	81
ADRB2	NM_174231.1	F: 5ʹ-TGATCGCTGTGGATCGCTAC-3ʹ R: 5ʹ-CCGGTACCAGTGCATCTGAA-3ʹ	149
ADRB3	NM_174232.2	F: 5ʹ-ACCTTCATTCTGTTCCTTCTG-3ʹ R: 5ʹ-CTGTGAGGTAGGTGTGTCTA-3ʹ	145
ARRB1	NM_174243.3	F: 5ʹ-TTCAACACAGCCCAGTACAA-3ʹ R: 5ʹ-GTAGACCTTGCAGAACGTAGAG-3ʹ	87
ARRB2	NM_001205277.2	F: 5ʹ-GGACCAGGGTCTTCAAGAAAT-3ʹ R: 5ʹ-CACTAGCACCACACCATCTAC-3ʹ	116
ATGL	NM_001046005.2	F: 5ʹ-TCTGCCTGCTGATTGCTATG-3ʹ R: 5ʹ-GCAGACATTGGCCTGGATAA-3ʹ	130
BARK1	NM_174710.2	F: 5ʹ-GAGATCTTCGACACGTACATCAT-3ʹ R: 5ʹ-CACCTGCTTCTTCACCAGAT-3ʹ	105
BARK2	NM_174500.2	F: 5ʹ-GCCTTCCACACTCCAGATAAA-3ʹ R: 5ʹ-AAACCGCATCTCCTTCTCAG-3ʹ	111
CEBPA	NM_176784.2	F: 5ʹ-ATCGACATCAGCGCCTACAT-3ʹ R: 5ʹ-GCCCGGGTAGTCAAAGTCG-3ʹ	141
FABP4	NM_174314.2	F: 5ʹ-GGAAAGTCAAGAGCATCGTAAAC-3ʹ R: 5ʹ-TGGCAGTGACACCATTCAT-3ʹ	141
FASN	NM_001012669.1	F: 5ʹ-CCATCCTTCTGACCAAGAAGTC-3ʹ R: 5ʹ-AGGTGACGCCTTTCTCTTTG-3ʹ	99
HSL	NM_001080220.1	F: 5ʹ-GAGACTGGCATCAGTGTGAC-3ʹ R: 5ʹ-TGCACGTCTAGGTTCTGAATG-3ʹ	113
MGLL	NM_001206681.1	F: 5ʹ-AAAGTTCTGAACCTCGTCCTG-3ʹ R: 5ʹ-GATGTCCACCTCCGTCTTATTC-3ʹ	87
PLIN1	NM_001083699.1	F: 5ʹ-CTGAAGGACACCATCTCCAC-3ʹ R: 5ʹ-CATACTCGGCAGTGTCTCTC-3ʹ	145
PPARG	NM_181024.2	F: 5ʹ-ATCTGCTGCAAGCCTTGGA-3ʹ R: 5ʹ-TGGAGCAGCTTGGCAAAGA-3ʹ	76
GAPDH	NM_001034034.2	F: 5ʹ-CTGCCCGTTCGACAGATAG-3ʹ R: 5ʹ-CTCCGACCTTCACCATCTTG-3ʹ	76
RPS9	NM_001101152.2	F: 5ʹ-GAGCTGGGTTTGTCGCAAAA-3ʹ R: 5ʹ-GGTCGAGGCGGGACTTCT-3ʹ	65
SDHA	NM_174178.2	F: 5ʹ-ACCTGATGCTTTGTGCTCTG-3ʹ R: 5ʹ-TCGTACTCGTCAACCCTCTC-3ʹ	106

ABHD5, abhydrolase domain containing-5; ADRB1, ADRB2, and ADRB3, β-adrenergic receptors-1, 2, and 3; ARRB1 and ARRB2, β-arrestin-1 and 2; ATGL, adipose triglyceride lipase; BARK1 and BARK2, β-adrenergic receptor kinase-1 and 2; CEBPA, CCATT/enhancer-binding protein alpha; FABP4, fatty acid-binding protein-4; FASN, fatty acid synthase; HSL, hormone-sensitive lipase; MGLL, monoacylglycerol lipase; PLIN1, perilipin; PPARG, peroxisome proliferator-activated receptor gamma; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; RPS9, ribosomal protein 9; SHDA, succinate dehydrogenase.

Statistical analysis

All the experiments were performed in triplicate for at least three independent experiments. Data are expressed as mean ± pooled SEM. Relative mRNA amounts in s.c. and i.m. preadipocytes and adipocytes were analyzed by two-factor analysis of variance (SuperAnova, Abacus Concepts, Inc., Berkeley, CA), with adipocyte source (s.c. vs. i.m.) and stage of differentiation (preadipocytes vs adipocytes) as main effects and the source x stage of differentiation as the interaction. β-Adrenergic receptor (ADRB) expression in s.c. and i.m. preadipocytes and adipocytes was analyzed by one-factor analysis of variance (SuperAnova). Effects of β-AA and LUB treatments in differentiated adipocytes were analyzed by one-factor analysis of variance (SuperAnova). When different (P ≤ 0.05), means were separated by Fisher’s Protected Least Squares Differences (SuperAnova).

Results

Induction of differentiation of bovine s.c. and i.m. preadipocytes

Before inducing adipogenic differentiation, s.c. and i.m. preadipocytes had fibroblast-like morphologies (Figure 1a and b). Small lipid droplets became visible under light microscopy by day 3 upon addition of differentiation media. The s.c. and i.m. preadipocytes further developed into medium and large lipid-filled adipocytes by day 9 of culture in maintenance media supplemented with 20 μg/mL insulin and 1 μM rosiglitazone (Figure 1c and d).

Figure 1.

Morphological changes after induction of differentiation in bovine s.c. and i.m. preadipocytes. (a) s.c. preadipocytes. (b) i.m. preadipocytes. (c) Differentiated s.c. adipocytes. (d) Differentiated i.m. adipocytes.

β-Adrenergic receptor and adipogenic gene expression

The source x stage of differentiation interaction was significant for ADRB1 (P = 0.005) and ADRB2 (P = 0.01) in that expression of ADRB1 and ADRB2 was greater in s.c. adipocytes than in s.c. preadipocytes, but expression of ADRB1 and ADRB2 did not change after differentiation of i.m. adipocytes (Table 2). Expression of ADRB3 did not change after differentiation of s.c. and i.m. preadipocytes to adipocytes (P = 0.20). CCATT/enhancer-binding protein alpha (CEBPA) expression increased after differentiation in s.c. and i.m. adipocytes (P = 0.006). The source x stage of differentiation interaction was significant for peroxisome proliferator-activated receptor gamma (PPARG) (P < 0.001) and fatty acid binding protein-4 (FABP4) (P = 0.004); expression of PPARG was greater in s.c. adipocytes than in s.c. preadipocytes, but expression of PPARG was the same in i.m. preadipocytes and adipocytes. FABP4 expression increased upon differentiation of both s.c. and i.m. adipocytes (P < 0.001), but FABP4 expression increased to a greater extent in s.c. adipocytes (source x stage of differentiation interaction P = 0.004).

Table 2.

Relative gene expression in subcutaneous and intramuscular preadipocytes and differentiated adipocytes¹

Item	Preadipocytes		Adipocytes		Pooled SEM	PREADIP vs ADIP	P-value	Interaction
	Subcutaneous	Intramuscular	Subcutaneous	Intramuscular			SC vs. IM
ADRB1	0.84b,y	1.38b,y	3.84a	1.71b	0.34	0.001	0.06	0.005
ADRB2	1.77bx	4.16abx	5.54a	2.38b	0.57	0.23	0.69	0.01
ADRB3	1.21x,y	1.72y	3.24	1.99	0.35	0.11	0.59	0.20
Pooled SEM	0.15	0.42	1.43	0.14
P-value	0.001	0.001	0.25	0.06
CEBPA	1.02	1.03	2.48	1.73	0.21	0.006	0.27	0.26
PPARG	1.08b	1.05b	6.53a	0.83b	0.63	< 0.001	< 0.001	< 0.001
FABP4	1.09c	1.05c	54.7a	15.0b	6.21	< 0.001	0.004	0.004

Values are mean ± SEM. n = 4 separate experiments. ADRB1, ADRB2, and ADRB3, β-adrenergic receptors-1, 2, and 3; CEBPA, CCATT/enhancer-binding protein alpha; FABP4, fatty acid-binding protein-4; PPARG, peroxisome proliferator-activated receptor gamma; PREADIP, preadipocytes; ADIP, differentiated adipocytes; SC, subcutaneous; IM, intramuscular.

Means within a row with common superscripts do not differ (P > 0.05).

Means within a column for ADRB with common superscripts do not differ (P > 0.05).

Adipocyte viability (MDH activity) after exposure to β-AR agonists and LUB

We evaluated the viability, measured as MDH activity, of differentiated s.c. adipocytes to determine the optimal concentration of β-AA and LUB for subsequent studies. Following 1 h incubation, DOB markedly increased MDH activity of s.c. adipocytes at concentrations of 10 µM (10⁻⁵ M) and 100 µM (10⁻⁴ M) (P < 0.05) (Figure 2). Treatment of s.c. adipocytes with SAL did not alter s.c. adipocyte MDH activity, but LUB reduced adipocyte MDH activity by 46% and 63% at concentrations of 10 and 100 µM, respectively (P < 0.001). Based on these results, we selected 10 µM DOB and SAL and 1 µM LUB for subsequent studies.

Figure 2.

Cell viability (% control) of differentiated s.c. adipocytes after 1 h treatment with 0 or 10⁻⁹ to 10⁻⁴ M dobutamine, salbutamol, or lubabegron fumarate. Data are expressed as means ± SEM (n = 4 separate experiment). *P ≤ 0.001, **P ≤ 0.0001, vs. control (either MeOH or DMSO).

Lubabegron fumarate as a β₁-and β₂-AR antagonist

In s.c. adipocytes, DOB increased cAMP and glycerol production and PKA activity, and cAMP production and PKA activity was strongly depressed by preincubation with LUB or PRO (P < 0.001) (Table 3). Glycerol production was not affected by preincubation with DOB but was depressed by PRO. SAL increased PKA activity but not cAMP or glycerol production in s.c. adipocytes and LUB and PRO attenuated the effect of SAL on PKA activity (P < 0.001). Treatment of i.m. adipocytes with SAL increased cAMP production and PKA activity, and this effect was attenuated by preincubation with LUB or PRO (P ≤ 0.006). Glycerol production was not affected by combinations of β-AA and LUB in i.m. adipocytes (P = 0.17).

Table 3.

Production of cAMP (pmol/mg protein), protein kinase A activity, and glycerol release into incubation medium (pmol glycerol/mg protein) in differentiated subcutaneous and intramuscular adipocytes treated with combinations of lubabegron fumarate, dobutamine, salbutamol, and propranolol¹

Item	Control	LUB	DOB	SAL	LUB/DOB	PRO/DOB	LUB/SAL	PRO/SAL	Pooled SEM	P-value
Subcutaneous adipocytes
cAMP	30bc	27bc	57a	45ab	23c	16c	17c	18c	3	< 0.001
PKA	75c	115c	258a	192b	97c	95c	93c	85c	11	< 0.001
Glycerol	31d	46cd	130a	53cd	100ab	72bc	31d	43cd	6	< 0.001
Intramuscular adipocytes
cAMP	23bc	27ab	20bc	34a	20bc	16c	18bc	18bc	1	0.002
PKA	93b	102ab	94b	123a	81b	95b	88b	78b	4	0.006
Glycerol	65	54	52	53	48	57	54	53	3	0.17

Values are mean ± SEM (n = 4 separate experiments). Control, basal; LUB, lubabegron fumarate, 1 µM; DOB, dobutamine, 10 µM; SAL, salbutamol, 10 µM; PRO, propranolol, 10 µM; DOB/LUB, pre-treated with LUB 5 min before addition of DOB; SAL/LUB, pre-treated with LUB 5 min before addition of SAL; DOB/PRO, pre-treated with PRO 5 min before addition of DOB; SAL/PRO, pre-treated with PRO 5 min before addition of SAL. Following pre-treatment with LUB or PRO, adipocytes were incubated an additional 15 min with the combinations of LUB, DOB, SAL, and PRO.

Within a tissue, means with common superscripts are not different (P > 0.05).

Expression of genes associated with lipolysis and adipogenesis

Adipose triglyceride lipase (ATGL) expression in s.c. adipocytes preincubated with LUB was less than for the other β-AA and LUB combinations (P = 0.01) (Table 4). For s.c. adipocytes, FABP4 expression was greatest in LUB + SAL-treated adipocytes and least in SAL-treated adipocytes (P < 0.001). DOB, SAL, LUB + DOB, and LUB + SAL increased hormone sensitive lipase (HSL) (P < 0.001) and perilipin (PLIN1) expression (P = 0.002), and LUB + SAL increased PPARG expression (P = 0.02) in s.c. adipocytes. Expression of abhydrolase domain containing-5 (ABHD5), fatty acid synthase (FASN), and monoacylglycerol lipase (MGLL) was not affected by β-AA or LUB (P ≥ 0.10) in s.c. adipocytes.

Table 4.

Relative expression of genes associated with lipid metabolism after 3 d treatment with combinations of lubabegron fumarate, dobutamine, and salbutamol in differentiated subcutaneous and intramuscular adipocytes¹

Item	Control	LUB	DOB	SAL	LUB/DOB	LUB/SAL	Pooled SEM	P-value
Subcutaneous adipocytes
ABHD5	1.00	1.17	1.37	1.33	1.44	1.41	0.07	0.10
ATGL	1.00ab	0.77b	1.12ab	1.01ab	1.54a	1.39a	0.09	0.01
FABP4	1.03bc	1.23b	1.15bc	0.94c	1.18b	1.79a	0.07	< 0.001
FASN	1.01	1.17	1.19	1.03	1.42	1.15	0.11	0.34
HSL	1.00d	1.94cd	2.76bc	4.18ab	4.66a	3.92ab	0.36	< 0.001
MGLL	1.01	1.09	1.26	1.23	1.28	1.32	0.07	0.23
PLIN1	1.00b	1.17b	1.90a	1.78a	1.77a	1.68a	0.10	0.002
PPARG	1.00b	1.06b	1.19b	1.12ab	1.23ab	1.56a	0.06	0.02
Intramuscular adipocytes
ABHD5	1.01	0.89	1.12	0.86	0.92	0.96	0.03	0.08
ATGL	1.01ab	0.90bc	1.12a	1.05ab	0.96abc	0.80c	0.03	0.001
FABP4	1.06b	1.54ab	1.76a	1.42ab	1.83a	1.50ab	0.09	0.01
FASN	1.02	1.18	0.84	0.99	1.11	0.84	0.06	0.10
HSL	1.00c	1.69abc	2.42a	1.59bc	1.94ab	1.26bc	0.13	0.001
MGLL	1.02	0.83	0.94	0.93	0.92	0.96	0.05	0.39
PLIN1	1.04	0.92	0.93	0.81	0.93	1.11	0.06	0.25
PPARG	1.02	0.78	0.93	0.79	0.91	0.93	0.06	0.31

Values are mean ± SEM (n = 3 separate experiments). Control, basal; LUB, lubabegron fumarate, 1 µM; DOB, dobutamine, 10 µM; SAL, salbutamol, 10 µM; DOB/LUB, combination of 10 µM DOB plus 1 µM LUB; SAL/LUB, combination of 10 µM SAL plus 1 µM LUB. At d 14 of differentiation, adipocytes were incubated with the above combinations of LUB, DOB, and SAL and allowed to incubate an additional 3 d.

Means with common superscripts are not different (P > 0.05). ABHD5, abhydrolase domain containing-5; ATGL, adipose triglyceride lipase; FASN, fatty acid synthase; FABP4, fatty acid-binding protein-4; HSL, hormone-sensitive lipase; MGLL, monoacylglycerol lipase; PLIN1, perilipin; PPARG, peroxisome proliferator-activated receptor gamma.

ATGL expression was depressed in i.m. adipocytes treated with LUB + SAL (P = 0.001), and DOB and LUB + DOB increased FABP4 expression (P = 0.01) and HSL expression (P = 0.001) above Control levels in i.m. adipocytes (Table 4). Relative mRNA levels of ABHD5, FASN, MGLL, PLIN1, and PPARG were unaffected by the combinations of β-AA and LUB (P ≥ 0.08) in i.m. adipocytes.

Expression of genes associated with β-AR and β-AR regulation

Expression of ADRB2 and ADRB3 responded similarly to β-AA and LUB in s.c. adipocytes, in that expression of those ADRB was greater than Control values in s.c. adipocytes incubated SAL or LUB + SAL (P ≤ 0.04) (Table 5); LUB + SAL also increased expression of ADRB1. Expression of β-arrestin-1 (ARRB1) was greater in s.c. adipocytes incubated with DOB or LUB + DOB than in adipocytes incubated with SAL (P = 0.04). Expression of ARRB2 and β-adrenergic receptor kinase-2 and -3 (GRK2 and GRK3, hereafter referred to BARK1 and BARK2) was not affected by β-AA or LUB (P ≥ 0.13).

Table 5.

Relative expression of genes associated with β-adrenergic receptors and receptor-desensitization after 3 d treatment with combinations of lubabegron fumarate, dobutamine, and salbutamol in differentiated subcutaneous and intramuscular adipocytes¹

Item	Control	LUB	DOB	SAL	LUB/DOB	LUB/SAL	Pooled SEM	P-value
Subcutaneous adipocytes
ADRB1	1.01b	1.53ab	1.44ab	1.78ab	1.78ab	1.98a	0.12	0.03
ADRB2	1.00b	1.35ab	1.35ab	1.80a	1.63ab	1.88a	0.12	0.04
ADRB3	1.01b	1.29ab	1.19ab	1.95a	1.70ab	2.08a	0.15	0.03
ARRB1	1.02ab	1.19ab	1.28a	0.82c	1.36a	1.21ab	0.07	0.04
ARRB2	1.05	1.00	0.99	1.29	1.45	1.40	0.10	0.28
BARK1	1.00	1.06	1.27	1.23	0.99	1.07	0.07	0.33
BARK2	1.01	1.11	1.23	1.00	0.82	1.02	0.06	0.13
Intramuscular adipocytes
ADRB1	1.06a	0.70b	0.92ab	0.84ab	0.70b	0.90ab	0.05	0.02
ADRB2	1.03a	0.64b	0.78ab	0.71b	0.65b	0.76b	0.04	0.01
ADRB3	1.07a	0.64b	0.76b	0.68b	0.71b	0.81ab	0.05	0.01
ARRB1	1.01	1.15	1.08	0.85	1.23	0.83	0.06	0.11
ARRB2	1.04a	0.74ab	0.61b	0.73ab	0.74ab	0.51b	0.06	0.02
BARK1	1.00a	0.85ab	0.89ab	0.88ab	0.93a	0.54b	0.05	0.02
BARK2	1.00a	0.97a	0.82ab	0.87a	0.85ab	0.59b	0.05	0.01

Values are mean ± SEM (n = 3 separate experiments). Control, basal; LUB, lubabegron fumarate, 1 µM; DOB, dobutamine, 10 µM; SAL, salbutamol, 10 µM; DOB/LUB, combination of 10 µM DOB plus 1 µM LUB; SAL/LUB, combination of 10 µM SAL plus 1 µM LUB. At d 14 of differentiation, adipocytes were incubated with the above combinations of LUB, DOB, and SAL and allowed to incubate an additional 3 d. ^abMeans sharing common superscripts are not different (P < 0.05). ADRB1, ADRB2, and ADRB3, β-adrenergic receptors-1, 2, and 3; ARRB1 and ARRB2, β-arrestin-1 and 2; BARK1 and BARK2, β-adrenergic receptor kinase-1 and 2.

Expression of ADRB1, ADRB2, and ADRB3 was depressed from Control values by LUB and LUB + DOB; SAL and LUB + SAL depressed ADRB2 expression from Control; and DOB and SAL depressed ADRB3 expression from Control in i.m. adipocytes (P ≤ 0.02) (Table 5). Expression of ARRB2, BARK1, and BARK2 was less in LUB + SAL-treated i.m. adipocytes than in Control adipocytes.

Discussion

We tested the apparent viability of differentiated s.c. adipocytes after exposure to DOB, SAL, and LUB with the primary goal of establishing optimal concentrations of LUB for subsequent experiments. Dobutamine strongly promoted the viability of adipocytes at 10 and 100 µM, whereas SAL did not affect adipocyte viability at any concentration. As indicated above, the CCK-8 assay measures MDH activity, which is taken as a measure of the number of viable cells. The nearly fourfold increase in apparent viability caused by 100 µM DOB suggests an increase in the number of viable s.c. adipocytes, but we instead interpret these results to mean that DOB increased MDH activity rather than causing an increase in viable s.c. adipocytes. We do not know the biological significance of these results. Regardless, the CCK-8 assay indicated that LUB may have been cytotoxic at concentrations of 10 and 100 µM, so this assay was useful for establishing a concentration of 1 µM LUB for subsequent experiments.

Differentiation of s.c. and i.m preadipocytes to adipocytes was confirmed by lipid filling and increased expression of adipogenic genes. Thus, CEBPA, PPARG, and FABP4 expression was greater in s.c. adipocytes than in s.c. preadipocytes, and CEBPA and FABP4 expression increased in i.m. adipocytes upon differentiation. The family of FABPs are transport proteins for fatty acids and FABP4 (more commonly called aP2) was isolated and the complementary DNA cloned from human adipose tissue (Baxa et al., 1989). We subsequently reported that FABP activity increased with marbling score in Angus steers (Moore et al., 1991). FABP4 expression increased in differentiated s.c. adipocytes exposed to LUB + SAL for 3 d, suggesting that LUB and this β₂-AA promoted s.c. adipocyte differentiation. In i.m. adipocytes, DOB (a β₁-AA) and LUB + DOB upregulated FABP4 expression. We interpret these results to mean that β₂-AA increase the capacity of s.c. adipocytes to transport fatty acids (via FABP4) resulting from elevated lipolysis, whereas β₁-AA would serve the same function in i.m. adipocytes. This is consistent with the acute stimulation of cAMP production and PKA activity; DOB more strongly elevated cAMP production and PKA activity in s.c. adipocytes and SAL more strongly stimulated these processes in i.m. adipocytes. It should be noted that there were no differences in expression levels of ADRB1, ADRB2, or ADRB3 in differentiated s.c. and i.m. adipocytes, and the protein amount of the ADRB was not measured in the study. At best, we can conclude that acute stimulation by β-AA increases aspects of the PKA signaling cascade, which were attenuated by the β-AR antagonist LUB. Also, chronic (3-d) treatment with β-AA increases expression of ADRB1-3, but this was not affected by LUB.

It is problematic to extrapolate results from cell culture studies to biological processes in adipose tissue explants or in the whole animal. For example, in the current study, expression levels of all three ADRB were similar in differentiated s.c. adipocytes, whereas, in s.c. adipose tissue ex vivo explants, the expression of ADRB2 was nearly fivefold greater than the expression of ADRB1 or ADRB3 (Hwang et al., 2021). Similarly, expression levels of ADRB1-3 were not statistically different in differentiated i.m. adipocytes but, in i.m. adipose tissue ex vivo explants, the expression of ADRB2 was nearly twofold greater than the expression of ADRB1 or ADRB3 (Hwang et al., 2021). Zheng et al. (2018) reported ADRB1-3 expression in bovine liver, muscle, and s.c. adipose tissue. ADRB2 expression was greater than ADRB1 expression and ADRB3 expression was barely detectable in s.c. adipose tissue, which contrasts with our previous (Hwang et al., 2021) and the current study, in which ADRB3 expression was readily detectable and was similar to ADRB1 expression.

ADRB3 expression has been demonstrated in tissues from several livestock species (Cousin et al., 1993; Casteilla et al., 1994; Mersmann, 1998; Johnson et al., 2014; Zheng et al., 2018) but, to the best of our knowledge, we are the first to report the expression of all three ADRB in i.m. adipose tissue (Hwang et al., 2021 and the current study). Zheng et al. (2018) reported expression ADRB in muscle, but they did not describe which muscle they used, or if i.m. adipose tissue was removed from the muscle samples before processing for RT–qPCR analysis. Casteilla et al. (1994) first reported ADRB3 gene expression in bovine brown adipose tissue (BAT), which increased markedly during the third trimester is calf perirenal BAT. We subsequently demonstrated a similar pattern of uncoupling protein-1 expression during the third trimester of gestation in bovine fetal perirenal BAT (Landis et al., 2002), and we interpreted these findings to mean that β₃-AR activity is important for BAT development and/or metabolic functionality.

The biological function of the β₃-AR in bovine white adipose tissue is unknown, and Hwang et al. (2021) reported conflicting results. We reported EC₅₀ values calculated based on cAMP accumulation in Chinese hamster ovary (CHO) cells, and LUB was highly selective for the β₃-AR, with an EC₅₀ of 6.0 × 10⁻⁹ M. The EC₅₀ of LUB for the β₁-AR in CHO was 1.9 × 10⁻⁷ M (i.e., the β₁-AR has less affinity to LUB), and there was no detectible agonistic activity for LUB at the β₂-AR. However, in that same report, cAMP and non-esterified fatty acid production was unaffected by LUB in either s.c. or i.m. adipose tissue explants (Hwang et al., 2021). Instead, LUB strongly attenuated SAL- and DOB-stimulated cAMP production (Hwang et al., 2021), indicating that CHO cells and adipose tissues obtained from live cattle responded quite differently to LUB. Importantly, in the current study, the β-AR antagonist efficacy of LUB was equivalent PRO in s.c. adipocytes, even though the concentration of LUB was 10-fold lower (1 µM) than that of PRO (10 µM). Thus, LUB is a powerful antagonist of β₁-AR and β₂-AR in bovine s.c. adipocytes.

The current study measured the expression of several genes associated with the regulation of TAG turnover and β-AR regulation. The activator protein ADHB5 strongly stimulates ATGL, which is responsible for the initial hydrolysis of TAG (Zimmermann et al., 2009). Activated PKA phosphorylates PLIN1 on the surface of lipid droplets and ABHD5 binds and activates ATGL to hydrolyze TAG. Activated PKA also phosphorylates HSL, after which HSL translocates to lipid droplets where it binds to PLIN1 to hydrolyze diacylglycerols. Subsequently, MGLL hydrolyzes monoacylglycerols to fatty acids and glycerol (Zimmermann et al., 2009; Zechner et al., 2012). Expression of ATGL was increased by LUB + DOB and LUB + SAL in s.c. adipocytes, and DOB, SAL, LUB + DOB, and LUB + SAL upregulated HSL expression. Taken together, the results of the current study indicate that β₁- and β₂-AA promote lipolysis in s.c. adipocytes and would be expected to promote lipolysis in cattle treated with β-AA.

Bryem et al. (1996) reported that infusion of cimaterol in steers elevated plasma nonesterified fatty acids (NEFA) concentration in a dose-dependent manner, suggesting that this early β-AA promoted lipolysis in adipose tissue. Prolonged stimulation of β-AR by β-AA triggers a rapid attenuation of receptor responsiveness, known as desensitization. Downregulation of the β-AR occurs in response to chronic agonist over-exposure of β-AA (Ferguson, 2001; Johnson, 2006; Dilger et al., 2021). O’Connor et al. (1991) demonstrated a 50% increase in plasma NEFA concentrations in young ram lambs within 2 h following feeding 20 ppm cimaterol. Plasma NEFA concentration declined sharply after 3 h post-feeding but remained at concentrations approximately threefold greater than baseline at 12 h post-feeding (their last sampling time point). The decline in plasma NEFA concentration following peak levels at 3 h suggests desensitization of the β-AR.

The G protein-coupled receptor kinases BARK1 and BARK2 phosphorylate activate β-AR on the 11 serine and threonine residues in the intracellular, C-terminal portion of β₂-AR, which promotes the binding of arrestin (encoded by ARRB) to the receptors. Arrestin binding to phosphorylated, activated β-AR prevents further receptor stimulation of the G protein-coupled complex, which causes β₂-AR desensitization and a concomitant reduction in cAMP production (Gurevich and Gurevich, 2019). The β₃-AR C-terminal loop is truncated, containing only three serine residues and lacking consensus sequences for PKA-mediated phosphorylation (Liggett et al., 1993), and thus would be less likely to be downregulated with chronic exposure to β-AA. Exposure of s.c. adipocytes to DOB and LUB + DOB for 3 d increased expression of ARRB1, suggesting that DOB would promote β₁-AR desensitization (which was not blocked by LUB). Conversely, SAL depressed expression of ARRB1 in s.c. adipocytes, suggesting that this β₂-AA would attenuate β₁-AR desensitization.

Data generated from i.m. adipocytes do not provide a clear picture of the effects of β-AA. Expression of ARRB2 was depressed in i.m. adipocytes treated for 3 d with DOB or SAL, and β-AA and LUB generally decreased ADRB1-3 expression. We can only speculate as to the effects of LUB on carcass adiposity and marbling score in beef cattle. For example, downregulation of ADRB1-3 expression by LUB should render i.m. adipose tissue less sensitive to β-AA in beef cattle that concurrently are administered ZIL or RAC. Conversely, downregulation of ARRB2 and BARK1 and BARK2 in i.m. adipocytes (as demonstrated in the current study) should attenuate β₂-AR desensitization in i.m. adipose tissue in vivo. Clearly, the net effect of co-treating beef cattle with LUB and a β-AA remains to be established.

At the time Hwang et al. (2021) was published, we did not have access to carcass data from cattle treated with LUB. Since that time, Kube et al. (2021) and Teeter et al. (2021) reported that LUB depressed beef marbling score and increased LM surface area in a dose-dependent manner. We recently reported that ZIL reduces marbling score through dilution of marbling by increased LM surface area and by a reduction in i.m. adipocyte volume (Gonzales et al., 2020). The mechanism by which ZIL reduced marbling score is unknown because treatment of s.c. and i.m. adipose tissue explants with ZIL had no effect on cAMP production or on glycerol or NEFA release (Hwang et al., 2021). LUB clearly is a β-AR antagonist (Hwang et al., 2021 and the current study), so it was unexpected that LUB and ZIL had such similar effects on marbling score. As a β-AR antagonist, LUB may have demonstrable biological effects only if cattle are concurrently challenged with a β-AA, e.g., if treated with ZIL and LUB at the same time. Marbling score primarily is depressed by increased dilution effect (Gonzales et al., 2020) and the reduction in marbling score could be caused by repartitioning of energy to support increased myofibrillar protein synthesis (Smith et al., 1989, 1995), so β-AA and LUB may not have a direct effect on lipolysis in vivo.

Hwang et al. (2021) and the current study demonstrated that s.c. and i.m. adipose tissue and adipocytes respond quite differently to β-AA and LUB. It is possible that s.c. and i.m. adipose tissues serve distinctly different functions in cattle. Subcutaneous adipose tissue serves as a readily mobilizable, systemic source of energy, whereas i.m. adipose tissue could serve primarily as fuel for adjacent muscle fibers. This is suggested by the greater amounts of marbling adipose tissue in cold-adapted cattle (Bos taurus, including Wagyu cattle) than is seen in warm-adapted cattle (Bos indicus). This is consistent with previous investigation from our laboratory (e.g., Smith and Crouse, 1984; Westbrook et al., 2021) that demonstrated that subcutaneous adipose tissue and marbling consist of metabolically distinct adipocytes.

In conclusion, we have demonstrated that LUB is a potent antagonist to β-AR₁ and β-AR₂ in bovine s.c. adipocytes cultured in vitro. Additionally, i.m. adipocytes were less sensitive than s.c. adipocytes to the specific β-AR₁ and β-AR₂ agonists, DOB and SAL.

Glossary

Abbreviations

BAT

brown adipose tissue

BSA

bovine serum albumin

cAMP

cyclic adenosine monophosphate

CHO

Chinese hamster ovary

CT

cycle threshold

DMEM/F12

Dulbecco’s modified Eagle’s Medium, nutrient mixture Ham’s F12

DMSO

dexamethasone

DOB

dobutamine hydrochloride

DPBS

Dulbecco’s phosphate-buffer saline

IBMX

3-isobutyl-1-methylxanthine

i.m.

intramuscular

ISO

isoproteronol

LM

longissimus muscle

LUB

lubabegron fumarate

MDH

mitochondrial dehydrogenase

PRO

propranolol hydrochloride

RAC

ractopamine hydrochloride

RPS9

ribosomal protein 9

RT–qPCR

realtime quantitative polymerase chain reaction

SAL

salbutamol sulfate

s.c.

subcutaneous

TAG

triacylglycerol

ZIL

zilpaterol hydrochloride

Funding

This article was supported by funds from Texas AgriLife Research.

Conflict of Interest Statement

The authors Hwang and Smith have no conflict of interest. Elanco Animal Health provided no funds to conduct this research, although Elanco Animal Health generously provided lubabegron fumarate.