Impact of high-fat diet on vasoconstrictor reactivity of white and brown adipose tissue resistance arteries

Abstract

Blood flow regulation is a critical factor for tissue oxygenation and substrate supply. Increased reactivity of arteries to vasoconstrictors may increase vascular resistance, resulting in reduced blood flow. We aimed to investigate the effect of a high-fat (HF) diet on stiffness and vasoconstrictor reactivity of white adipose tissue (WAT) and brown adipose tissue (BAT) resistance arteries and also investigated the interconversion of both adipose depots in the setting of a HF diet. Vasoconstrictor reactivity and passive morphology and mechanical properties of arteries from B6D2F1 mice (5 mo old) fed normal chow (NC) or a HF diet (8 wk) were measured using pressure myography. Receptor gene expression in WAT and BAT arteries and markers of WAT and BAT were assessed in whole tissue lysates by real-time RT-PCR. Despite greater receptor-independent vasoconstriction (in response to KCl, P < 0.01), vasoconstriction in response to angiotensin II (P < 0.01) was lower in NC-BAT than NC-WAT arteries and similar in response to endothelin-1 (P = 0.07) and norepinephrine (P = 0.11) in NC-BAT and NC-WAT arteries. With the exception of BAT artery reactivity to endothelin-1 and angiotensin II, the HF diet tended to attenuate reactivity in arteries from both adipose depots and increased expression of adipose markers in BAT. No significant differences in morphology or passive mechanical properties were found between adipose types or diet conditions. Alterations in gene expression of adipose markers after the HF diet suggest beiging of BAT. An increase in brown adipocytes in the absence of increased BAT mass may be a compensatory mechanism to dissipate excess energy from a HF diet.

NEW & NOTEWORTHY Despite no differences in passive mechanical properties and greater receptor-independent vasoconstriction, receptor-mediated vasoconstriction was either lower in brown than white adipose tissue arteries or similar in brown and white adipose tissue arteries. A high-fat diet has a greater impact on vasoconstrictor responses in white adipose tissue but leads to altered adipose tissue gene expression consistent with beiging of the brown adipose tissue.

Listen to this article's corresponding podcast at https://ajpheart.podbean.com/e/vasoconstriction-in-white-and-brown-adipose/.

INTRODUCTION

Adipose tissue (AT) is a highly vascularized endocrine organ that plays a major role in energy homeostasis. Each of the two major forms of AT in mammals, white AT (WAT) and brown AT (BAT), has distinct functions (42). While a primary function of WAT is storage of excess energy as triglycerides, BAT dissipates a large amount of energy as heat (40). Although BAT was previously believed to be absent in adult humans, recent studies have demonstrated intact BAT within the aortic perivascular fat and in the subclavicular regions (41). In contrast to WAT, an increase in the amount or activation of BAT has been linked to improved metabolic function (19, 29).

Consumption of a high-fat (HF) diet is a primary risk factor for metabolic and vascular diseases, such as diabetes, hypertension, and coronary artery disease (23). HF diet can increase adiposity in humans and animal models, alter the levels and actions of adipokines and lipids, and induce macrophage infiltration of AT (1, 9, 26). A HF diet results in reduced expression of genes involved in lipid metabolism and adipocyte differentiation and increased expression of inflammatory markers (22, 31). Although still controversial, recent studies have also suggested that a HF diet can alter the expression of markers specific to WAT or BAT in both types of AT, a process termed “beiging” (14).

As in all other tissues, AT blood flow (ATBF) is tightly regulated (11). Impaired ATBF and hypoxia are associated with obesity and insulin resistance (11, 53). One likely explanation for these observations is that obesity is associated with increased peripheral vascular resistance, which is regulated by the balance of local vasoconstrictors and vasodilators. It is well established that endothelium-dependent dilation is impaired in obesity (9, 44, 52). Although reductions in vasodilator reactivity (9) have been demonstrated in WAT, little is known about the vasoreactivity of arteries in BAT or how this is impacted by diet-induced obesity. In addition, there is a dearth of information regarding the vasoconstrictor reactivity of arteries from either adipose depot in the setting of a HF diet. Inasmuch as ATBF and subsequent tissue oxygenation are regulated by the combined effects of endogenous vasodilators and vasoconstrictors, alterations in the vasoconstrictor reactivity of arterioles excised from these two distinct adipose depots likely have important implications for systemic metabolic function.

The aims of our study were 1) to compare the vasoconstrictor reactivity of the arteries in response to angiotensin II (ANG II), endothelin (ET)-1, norepinephrine (NE), and KCl; 2) to determine if this reactivity is differentially affected after a HF diet; 3) to determine if passive mechanical properties of the resistance arteries differ between adipose depots and how they are impacted by a HF diet, and 4) to investigate whether a HF diet preferentially impacts the beiging of these two AT depots.

MATERIALS AND METHODS

Animals

Young (8-wk-old) male B6D2F1 mice were obtained from Charles River. This mouse model, which has been studied extensively for metabolic assessment by our group (9), has demonstrated reduced insulin sensitivity, accompanied by elevated tissue inflammation, after a HF diet. All mice were housed for 3 mo on a 12:12-h light-dark cycle at the Salt Lake City Veterans Affairs Medical Center Animal Facility. Mice were fed normal rodent chow (NC; 16% kcal from fat, 55% kcal from carbohydrate, and 29% kcal from protein, no. 8640, Harlan Teklad-standard rodent chow) or a commercially available HF diet [41% kcal from fat (41% kcal from saturated/total fat), 41% kcal from carbohydrate, and 18% kcal from protein, Harlan adjusted-fat diet TD.96132] and housed in standard mouse cages for 8 wk before they were euthanized, as previously described (27). Access to food and water was provided ad libitum. Euthanasia was performed by exsanguination via cardiac puncture under isoflurane anesthesia. Before anesthesia for euthanasia, blood (~5 μl) was collected from a tail nick while the mouse was handheld, and blood glucose was measured using a portable glucose meter. All animal procedures conformed with the National Institutes of Health Guide for the Care and Use of Laboratory Animals (version 8, Revised 2011) and were approved by the Salt Lake City Veterans Affairs Medical Center Animal Care and Use Committee and the Institutional Animal Care and Use Committee of the University of Utah.

Vasoconstrictor Reactivity

Vasoconstrictor reactivity was assessed in arteries excised from epididymal WAT and subscapular BAT, as previously described (8, 24). Briefly, AT was placed in cold (4°C) physiological salt solution (PSS), and resistance-sized (i.e., <300-µm maximal lumen diameter) arteries were dissected and placed in the chamber of a pressure myograph (DMT, Atlanta, GA) containing PSS. Arteries were cannulated onto glass pipettes, pressurized to 50-mmHg intraluminal pressure, and allowed to equilibrate for 1 h at 37°C. Changes in lumen diameter in response to the cumulative addition of ANG II (10⁻¹⁰–10⁻⁴ log M), ET-1 (10⁻¹¹–10⁻⁷ log M), and NE (10⁻⁹–10⁻⁴ log M) were visualized and assessed using MyoVIEW software (DMT). In addition, a concentration response to cumulative addition of KCl (10⁻⁶–10⁻¹ log M) was also determined to assess nonreceptor-mediated smooth muscle reactivity. Three 20-min washes with PSS were performed between each concentration response. Spontaneous tone and constrictor reactivity were calculated as a percentage of possible vasoconstriction, as previously described (10).

Passive Mechanical Properties

Morphological characteristics and passive mechanical properties of WAT and BAT arteries from NC- and HF diet-fed mice were assessed in isolated arteries, as previously described (24). Briefly, arteries were incubated in Ca²⁺-free PSS for ≥1 h before lumen diameters were measured in response to increasing [5–100 cmH₂O (3.7–73.5 mmHg)] intraluminal pressure in 5-cmH₂O increments. At the end of this passive pressure-diameter assessment, intraluminal pressure was returned to 68 cmH₂O, lumen diameter and wall thickness were measured, and the wall-to-lumen ratio was calculated. Circumferential stress and stretch were calculated from pressure-diameter responses, as previously described (24).

Vasoconstrictor Receptor Gene Expression

The remaining WAT and BAT arteries were excised from the respective adipose depots, cleared of surrounding adipocytes, and frozen in liquid nitrogen. Frozen arteries were homogenized, and total RNA was extracted using the miRNeasy Mini kit (Qiagen). Equal amounts of RNA were reverse transcribed with the QuantiTect reverse transcription kit (Qiagen). An additional genomic DNA digestion step was also used. Real-time PCR was performed using SYBR green PCR master mix (Qiagen) and the iCycler PCR system (Bio-Rad). The target genes tested were as follows: ANG II receptor type 1a (Agtr1a), ANG II receptor type 2 (Agtr2), adrenergic receptor-β₃ (Adrβ3), adrenergic receptor-α₁ (Adrα1), adrenergic receptor-α₂ (Adrα2), ET-1 (Et-1), ET receptor type A (Ednra), and ET receptor type B (Ednrb). Primer sequences are shown in Table 1. Threshold cycle (C_T) values of target genes were normalized to the C_T value for 18S [mouse 18S primer (catalog no. QT02448075, Qiagen)]. PCR was carried out with the iQ5 real-time PCR detection system (Bio-Rad).

Table 1.

Primer sequences used for gene expression analysis

	Primers
Agtr1a
Forward	5′-CACTCAAGCCTGTCTACGAAA-3′
Reverse	5′-CACTCCACCTCAGAACAAGAC-3′
Adrβ3
Forward	5′-ACAGGAATGCCACTCCAATC-3′
Reverse	5′-GAGCATAGACGAAGAGCATCAC-3′
Adrα1
Forward	5′-CAGATGGAGTCTGTGAATGGAA-3′
Reverse	5′-AATGGTTGGAACTTGGTGATTT-3′
Adrα2
Forward	5′-TTCTTTTTCACCTACACGCTCA-3′
Reverse	5′-TGTAGATAACAGGGTTCAGCGA-3′
ET-1
Forward	5′-CTTCCCAATAAGGCCACAGACCAG-3′
Reverse	5′-AGCCACACAGATGGTCTTGCTAAG-3′
Ednra
Forward	5′-AATCATTGTGGTCGAAAGGC-3′
Reverse	5′-GGCCCTTGGAGACCTTATCT-3′
Ednrb
Forward	5′-AGATATCGAGCTGTTGCTTCTT-3′
Reverse	5′-CCACAGAGACCACCCAAATTA-3′
Agtr2
Forward	5′-GGACTCATTGGTGCCAGTTG-3′
Reverse	5′-GCTTACTTCAGCCTGCATTT-3′
Retn
Forward	5′-CAGAAGGCACAGCAGTCTTGA-3′
Reverse	5′-CTGTCCAGTCTATCCTTGCACAC-3′
Pltp
Forward	5′-GTCTAAAATGAATATGGCCTTCG-3′
Reverse	5′-CCAGAAGTGATGAACGTGGA-3′
Gstα3
Forward	5′-CGCTTTCAGGAGAGGGAAGTTG-3′
Reverse	5′-AGGAACAAACCAGGAACCGTTAC-3′
Serpina3k
Forward	5′-AGCCAACAACCCTGAACATC-3′
Reverse	5′-TCCCCATAGCTACAATGAAGG-3′
Ucp1
Forward	5′-ACTGCCACACCTCCAGTCATT-3′
Reverse	5′-CTTTGCCTCACTCAGGATTGG-3′
Elovl3
Forward	5′-TCCGCGTTCTCATGTAGGTCT-3′
Reverse	5′-GGACCTGATGCAACCCTATGA-3′
Cidea
Forward	5′-TGCTCTTCTGTATCGCCCAGT-3′
Reverse	5′-GCCGTGTTAAGGAATCTGCTG-3′

Agtr1a, ANG II receptor type 1a; Adrβ3, adrenergic receptor-β₃; Adrα1, adrenergic receptor-α₁; Adrα2, adrenergic receptor-α₂; ET-1, endothelin-1; Ednra, endothelin receptor type A; Ednrb, endothelin receptor type B; Agtr2, ANG II receptor type 2; Retn, resistin; Pltp, phospholipid transfer protein; Gstα3, glutathione S-transferase-α₃; Serpina3k, serine (or cysteine) protease inhibitor A3K precursor; Ucp1, uncoupling protein 1; Elovl3, fatty acid elongase 3; Cidea, cell death-inducing DNA fragmentation factor-α-like effector A.

WAT and BAT Marker Gene Expression

WAT and BAT tissues were weighed and snap frozen in liquid nitrogen. Total RNA was extracted using the RNeasy Mini kit (Qiagen), and cDNA was transcribed using a cDNA synthesis kit (Qiagen) according to the manufacturer’s instructions. Gene expression for WAT-specific markers [resistin (Retn) (3), phospholipid transfer protein (Pltp), glutathione S-transferase-α₃ (Gsta3) (47), and serine (or cysteine) protease inhibitor A3K precursor (Serpina3k) (15)] as well as for BAT-specific markers [uncoupling protein 1 (Ucp1), cell death-inducing DNA fragmentation factor-α-like effector A (Cidea1) (15), and fatty acid elongase 3 (Elovl3) (39)] was determined by real-time RT- PCR. Primer sequences are shown in Table 1. RNA was quantified as described above.

Statistics

For animal and vessel characteristics, group differences were determined by one-way ANOVA. Group differences in the concentration-response and pressure-diameter relationships were determined by repeated-measures ANOVA using SPSS software (version 19, SPSS, Armonk, NY). Least-squares difference (LSD) post hoc analyses were performed when significant interactions or group differences were found. P values reported in results are for group differences from the repeated-measures ANOVA or for LSD post hoc analyses where appropriate. EC₅₀ values were calculated by fitting each concentration response to a four-parameter logistic equation. EC₅₀ values for BAT and WAT responses to ANG II were calculated for only the constriction portion of the curve, inasmuch as the complete response does not fit a logistic equation. Values are means ± SE. Significance was determined as P ≤ 0.05.

RESULTS

Body and AT Mass of Experimental Mice

Total body mass was greater (P < 0.01) and fed blood glucose was higher (P < 0.01) in HF diet-fed compared with NC-fed mice, although kilocalories consumed per day (P = 0.15) did not differ between groups (Table 2). The HF diet was also associated with a nearly twofold increase in epididymal WAT mass (P < 0.01), accounting for 4.4% of total body mass compared with only 2.4% in NC-fed mice (Table 2). Subscapular BAT mass was unaltered by the HF diet (P = 0.21; Table 2).

Table 2.

Characteristics of mice fed NC or HF diet

	Diet
	NC (n = 10)	HF (n = 15)
	Mice
Body mass, g	32 ± 1	35 ± 1*
Blood glucose, mg/dl	119 ± 6	156 ± 6*
Food intake, kcal/day	13 ± 0.3	14 ± 0.6
	Tissue
Epididymal white adipose tissue
Weight, g	0.77 ± 0.10	1.6 ± 0.08*
%Body mass	2.4 ± 0.26	4.42 ± 0.15*
Brown adipose tissue
Weight, g	0.25 ± 0.03	0.33 ± 0.02
%Body mass	0.79 ± 0.09	0.92 ± 0.05

Values are means ± SE; n, number of mice. NC, normal chow; HF diet, high-fat diet.

^*P < 0.05 vs. NC.

Receptor-Independent Vasoconstrictor Capacity

In response to KCl, a receptor-independent agonist, vasoconstriction was greater in BAT than WAT arteries from NC-fed mice (P < 0.01). The HF diet had no effect on vasoconstriction to KCl in arteries from either adipose depot (P = 0.40 for WAT and P = 0.24 for BAT; Fig. 1A). Sensitivity of excised arteries to KCl was not affected by depot (WAT vs. BAT) or diet (NC vs. HF diet; Table 3).

Fig. 1.

Vasoconstriction in isolated arteries from white adipose tissue (WAT) and brown adipose tissue (BAT) of normal chow (NC)- and high-fat (HF) diet-fed male B6D2F1 mice. A–D: concentration responses to KCl, angiotensin II, norepinephrine, and endothelin-1 were assessed using a pressure myograph. Values are means ± SE; n = 5–6/group. *P ≤ 0.05, group difference from WAT within diet; †P ≤ 0.05, difference from NC within depot (by repeated-measures ANOVA).

Table 3.

Characteristics and receptor gene expression of AT excised from epididymal WAT and subscapular BAT from mice fed NC and HF diets

	WAT		BAT
	NC	HF diet	NC	HF diet
	Characteristics
Maximal diameter, µm	181 ± 5	197 ± 10	179 ± 23	163 ± 9
Wall thickness, µm	21 ± 2	19 ± 2	20 ± 1	20 ± 1
Wall-to-lumen ratio	0.12 ± 0.01	0.10 ± 0.01	0.11 ± 0.01	0.12 ± 0.01
Spontaneous tone, %	1.59 ± 1.37	1.48 ± 0.66	7.86 ± 3.46	6.47 ± 3.46
EC50, log M
ANG II	−8.09 ± 0.56	−8.55 ± 0.24	−8.19 ± 0.13	−8.29 ± 0.77
Norepinephrine	−6.65 ± 0.30	−6.24 ± 0.33	−6.02 ± 0.28	−4.98 ± 0.79
Endothelin-1	−9.67 ± 0.28	−9.05 ± 0.32	−8.78 ± 0.32	−9.38 ± 0.27
			(P = 0.06)
KCl	−3.45 ± 0.43	−3.35 ± 0.38	4.29 ± 0.29	−4.24 ± 0.25
	Receptor expression
Agtr1a	1 ± 0.26	0.85 ± 0.27	3.58 ± 1.2	0.95 ± 0.37
Adrβ3	1 ± 0.24	0.65 ± 0.16	0.36 ± 0.15*	0.63 ± 0.31
Adrα1	1 ± 0.13	1.67 ± 0.47	4.87 ± 1.26*	7.67 ± 2.35†
Adrα2	1 ± 0.18	1.05 ± 0.11	0.13 ± 0.07*	0.08 ± 0.03†
Ednra	1 ± 0.12	0.81 ± 0.12	0.09 ± 0.02*	0.11 ± 0.03
Ednrb	1 ± 0.13	0.76 ± 0.13	0.09 ± 0.02*	0.10 ± 0.03

Values are means ± SE. Receptor expression values were calculated by measurement of $2^{(-\Delta {\text{C}}_{\text{T}})}$, where C_T is threshold cycle, and normalized to mean values of the test gene in white adipose tissue (WAT) in normal chow (NC)-fed mice. BAT, brown adipose tissue; ANG II, angiotensin II; Agtr1a, ANG II receptor type 1a; Adrβ3, adrenergic receptor-β₃; Adrα1, adrenergic receptor-α₁; Adrα2, adrenergic receptor-α₂; Ednra, endothelin receptor type A; Ednrb, endothelin receptor type B.

^*P < 0.05 vs. NC-WAT;

^†P < 0.05 vs. HF-WAT.

ANG II-Induced Vasoconstriction and Receptor Gene Expression

In NC-fed mice, ANG II induced a greater vasoconstriction in WAT than BAT arteries, with the maximum vasoconstriction ~80% lower in BAT than WAT arteries (P < 0.01). The effect of ANG II on WAT arteries was biphasic: arteries responded with vasoconstriction to lower concentrations (10⁻¹⁰–10⁻⁷ log M) followed by relaxation at higher concentrations (10⁻⁷–10⁻⁴ log M) of ANG II (Fig. 1B). The HF diet was associated with a reduction in the vasoconstrictor portion of the ANG II concentration response in WAT arteries, with maximal vasoconstriction 43% lower in arteries from HF diet-fed than NC-fed mice (P < 0.01, NC-WAT vs. HF-WAT, both concentration response and maximal vasoconstriction). Irrespective of diet, BAT arteries demonstrated little vasoreactivity to ANG II (Fig. 1B). Additionally, there were no group differences in sensitivity to ANG II for either artery (EC₅₀, P < 0.93; Table 3). Despite a lack of vasoreactivity to ANG II in BAT arteries, gene expression for Agtr1a tended to be higher (P = 0.20) in BAT than WAT arteries in NC-fed mice (Table 3), whereas expression of Agtr2 was below the detection level of our system in all arteries. There was no effect of HF diet on receptor expression in WAT or BAT arteries.

Adrenergic Vasoreactivity and Receptor Gene Expression

NE-mediated vasoconstriction (Fig. 1C) and sensitivity (Table 3) did not differ between WAT and BAT arteries from NC-fed mice. Vasoconstriction to NE was lower in WAT (concentration response and maximal vasoconstriction: both P < 0.01) and BAT (concentration response and maximal vasoconstriction: both P < 0.06) arteries from HF diet-fed animals (Fig. 1C), but sensitivity was not impacted (Table 3). Although arterial gene expression for the adrenergic receptors differed between WAT and BAT, such that Adrα1 was higher (P < 0.01) and Adrα2 was lower (P < 0.01) in BAT than WAT of NC-fed mice, the HF diet did not have a significant effect on receptor expression in arterial tissue from either depot (Table 3).

ET-1-Mediated Vasoconstriction and Receptor Gene Expression

ET-1-mediated vasoconstriction was not different between WAT and BAT arteries from NC-fed mice (Fig. 1D). HF diet-fed mice exhibited ~30% lower (P < 0.01) maximal ET-1-mediated vasoconstriction in WAT arteries and tended (P = 0.10) to vasoconstrict less to ET-1 in BAT arteries (Fig. 1D). Maximum vasoconstriction and sensitivity to ET-1 were similar between diet groups in BAT arteries (both P > 0.50; Table 3). Although expression of Edn1 (1 ± 0.2 in NC-WAT and 0.36 ± 0.1 in NC-BAT, P = 0.01) and ET-1 receptors (Ednra and Ednrb) was higher in WAT than BAT arteries in NC-fed mice (P < 0.001), the HF diet did not affect receptor expression in arteries from either adipose depot (Table 3).

Passive Mechanical Properties of Adipose Arteries

There were no differences in maximal lumen diameter, wall thickness, or wall-to-lumen ratio between WAT and BAT arteries or between NC- and HF diet-fed mice (all P > 0.32; Table 3). Likewise, there were no group differences in the pressure-diameter response (P = 0.19; Fig. 2A) or circumferential stress or stretch (both P > 0.33; Fig. 2B) between adipose depots or diet conditions. Although there was a significant interaction by repeated-measures ANOVA for circumferential stress (P = 0.04), LSD post hoc analyses failed to find significant differences in circumferential stress between adipose depots or diet conditions (all P > 0.14).

Fig. 2.

Biomechanical properties of white adipose tissue (WAT) and brown adipose tissue (BAT) arteries of normal chow (NC)- and high-fat (HF) diet-fed male B6D2F1 mice. A: passive pressure-diameter relationship. B: circumferential stress-stretch relationship. Values are means ± SE; n = 4–8/group.

WAT and BAT Marker Expression in Adipose Depots

As expected, expression of WAT-specific markers (Retn, Pltp, Serpina3k, and Gsta3) was higher (P < 0.01) in WAT than BAT in NC-fed mice (Fig. 3) and expression of BAT-specific markers (Ucp1 and Cidea) was higher (P < 0.01) in BAT than WAT (Fig. 4). The HF diet did not affect expression of WAT markers in the epididymal WAT depot. However, expression of all WAT markers tested was higher in BAT isolated from HF diet-fed than NC-fed mice (P < 0.01). Although the HF diet did not affect expression of BAT markers in WAT, expression of two BAT markers, Ucp1 (P = 0.03) and Elovl3 (P = 0.02), was also higher in BAT isolated from HF diet- than NC-fed mice (Fig. 4).

Fig. 3.

Gene expression of white adipose tissue (WAT) markers in excised arteries from WAT and brown adipose tissue (BAT) of normal chow (NC)- and high-fat (HF) diet-fed male B6D2F1mice. A–D: mRNA expression of resistin (Retn), phospholipid transfer protein (Pltp), serine (or cysteine) protease inhibitor A3K precursor (Serpina3k), and glutathione S-transferase-α₃ (Gsta3) was assessed by real-time RT-PCR. Values are means ± SE; n = 4–6/group. *P < 0.05 vs. WAT within diet group; †P < 0.05 vs. NC within depot.

Fig. 4.

Gene expression of thermogenic, brown adipose tissue (BAT) markers in excised arteries from white adipose tissue (WAT) and BAT of normal chow (NC)- and high-fat (HF) diet-fed male B6D2F1 mice. A–C: mRNA expression of uncoupling protein 1 (Ucp1), cell death-inducing DNA fragmentation factor-α-like effector A (Cidea), and fatty acid elongase 3 (Elovl3) was assessed by real-time RT-PCR. Values are means ± SE; n = 5–6/group. *P < 0.05 vs. WAT within diet group; †P < 0.05 vs. NC within depot.

DISCUSSION

The major findings of this study are as follows. First, despite greater receptor-independent KCl-mediated vasoconstriction, vasoconstriction in response to ANG II was lower in BAT than WAT arteries but was similar in response to ET-1 and NE in BAT and WAT arteries. Second, the HF diet was associated with lower vasoconstriction in arteries from both AT depots, with the exceptions of reactivity to ANG II and ET-1 in BAT arteries. Third, differences in the passive mechanical properties of the arteries cannot explain the differences in vasoconstrictor reactivity. Finally, HF diet leads to an increase in expression of specific WAT and BAT markers in BAT, suggestive of preferential beiging of this tissue.

Adipose Artery Vasoconstrictor Reactivity

KCl.

In the present study, we observed greater vasoconstrictor reactivity to the nonreceptor-mediated vasoconstrictor KCl in BAT than WAT arteries of NC-fed mice. As KCl causes direct smooth muscle depolarization, leading to smooth muscle contraction via Ca²⁺ entry through voltage-sensitive Ca²⁺ channels, this finding suggests greater smooth muscle contractility in the arteries of BAT than WAT, which may result from differences in the intrinsic properties of the smooth muscle cells residing in the white and brown adipose arteries (13, 50), a possibility that would need further investigation.

ANG II.

We found that ANG II induces a biphasic response in arteries from the WAT: vasoconstriction at lower concentrations followed by relaxation at higher concentrations. Reports in the literature have suggested that repeated stimulation with ANG II can result in tachyphylaxis and arterial relaxation, and such a loss of tone may mediate the vasodilator portion of the ANG II response observed in the present study (7, 46). However, we previously demonstrated a biphasic response to ANG II in skeletal muscle arterioles (37) that was similar to our observation in WAT arteries. Importantly, the vasodilation to high concentrations of ANG II in skeletal muscle arterioles was mediated by active endothelium-dependent dilation, as it was abolished by removal of the endothelium (37). Furthermore, coronary arterioles have also been demonstrated to exhibit vasodilation at higher ANG II concentrations (10 nmol/l–1 µmol/l), although no vasoconstriction was observed at a lower concentration (0.1 nmol/l) in these arteries (55). This biphasic response may result in ANG II type 1 receptor activation of smooth muscle cells at low concentrations (37) and activation of ANG II type 1 receptors and ANG II type 2 receptors on the endothelium at higher concentrations (4, 5, 30). We believe that the present data are in agreement with previous studies suggesting an active biphasic ANG II response, rather than a tachyphylaxis, because lumen diameter was measured after vessels achieved a steady state at each concentration and because WAT arteries from NC-fed mice maintained ~30% constriction, even at the highest dose of ANG II, indicating that there was still a considerable amount of tone in the arteries. However, as we did not perform experiments after nitric oxide synthase blockade or denudation of the arteries, we cannot rule out tachyphylaxis as the cause of the relaxation to high doses of ANG II.

Interestingly, although we observed little responsiveness to ANG II in arteries isolated from BAT and no effect of the HF diet on this response, we found robust vasoconstriction to low concentrations of ANG II in arteries isolated from WAT that was blunted in arteries from HF diet-fed mice. Interestingly, previous studies have demonstrated an association of obesity-associated reductions in ATBF in the subcutaneous adipose depot with increased systemic release of ANG II (6). Additionally, increased plasma concentrations of and enhanced reactivity to ANG II have been reported in the brachial artery of obese individuals (35) and animal models of diet-induced obesity (33). Together, these results suggest that ANG II may have more profound effects on blood flow in WAT than BAT. Furthermore, in contrast to subcutaneous AT, the potential impact of pathological/physiological increases in ANG II to reduce blood flow (16) may be blunted in the visceral adipose depot after HF diet feeding. Since there is differential innervation of the sympathetic nervous system and differential sympathetic drive in visceral and subcutaneous adipose depots (2), we speculate that this may underlie differences in vasoconstrictor responses between depots in response to ANG II.

Norepinephrine.

AT is innervated by the sympathetic nervous system, whereby sympathetic outflow reduces ATBF. Despite significant differential expression of adrenergic receptors between adipose depots, i.e., increased expression of α₁-receptors and decreased expression of α₂-receptors in BAT arteries compared with WAT arteries, we found robust and similar vasoconstriction to NE in arteries from brown and white adipose depots. Because sensitivity to NE was similar between arteries from both adipose depots, an explanation may be alterations in receptor sensitization rather than receptor density (45), possibilities requiring further elucidation. In addition, in the present study, we observed a blunted NE-induced vasoconstrictor response, without concurrent changes in sensitivity or adrenergic receptor expression, in arteries from WAT and BAT of HF diet-fed mice. Furthermore, blunted vasoconstrictor responses could also be a compensatory mechanism of the resistance arteries to maintain homeostasis by controlling the balance between blood flow and metabolite clearance in the setting of chronic sympathetic nervous system activation, as in obesity or in the setting of a HF diet. Future studies are needed to determine the contribution of α- and β-receptors in modulating differential changes in reactivity of adipose arteries.

Endothelin-1.

In the present study, we found no difference in vascular reactivity to ET-1 between arteries isolated from WAT and BAT depots of NC-fed mice. However, ET-1-mediated vasoconstriction was lower in WAT arteries excised from HF than NC diet-fed mice. Previously, it has been demonstrated that the vasoconstrictor response to ET-1 was significantly higher (indicated by reduced forearm blood flow) in overweight and obese than normal weight adults (51), and while our observation is not in agreement with this finding, the discrepancy may result from tissue specificity in vasoconstrictor responsiveness. Indeed, while this previous report would suggest reduced limb blood flow in the presence of ET-1, our data suggest that blood flow to WAT may be maintained or enhanced in obesity, even in the presence of elevated ET-1.

Although we did not observe an effect of the HF diet on expression of ET-1 receptors of WAT or BAT arteries, there was a tendency (P = 0.09) for increased ET-1 expression in WAT arteries from HF diet-fed mice compared with NC-fed mice. Inasmuch as obesity is generally associated with hyperinsulinemia and insulin stimulates production and release of ET-1 (21), it is possible that the reduced vasoconstrictor response after a HF diet compensates for increased release of ET-1 from WAT arteries. Additionally, despite higher expression of ET receptor types A and B (Ednra and Ednrb) and ET-1 in WAT than BAT of NC-fed mice, there was no difference in vascular reactivity between depots. However, we found a tendency for reduced sensitivity to ET-1 in WAT compared with BAT arteries (P = 0.06; Table 3), and this lower sensitivity of the ET receptors to ET-1 could explain the lack of difference in reactivity in arteries from WAT compared with BAT.

Integrated responses.

We have found that, in response to physiologically relevant vasoconstrictors, WAT arteries are capable of robust vasoconstriction and that the HF diet is associated with reduced reactivity of these vessels to ANG II, NE, and ET-1. Because all these vasoconstrictors are induced by a HF diet/obesity, the interaction of these factors will likely contribute to alterations in vascular tone and blood flow. In contrast, although constriction of BAT arteries to NE and ET-1 is similar to our finding of constriction of WAT arteries, there was almost no response to ANG II, despite measurable mRNA for the relevant receptors. In BAT arteries, the HF diet had the greatest impact on responsiveness to NE, resulting in decreased reactivity. One possible explanation for the impaired NE-induced vasoreactivity in vivo is reduced vessel tone caused by release of nitric oxide from leukocytes that infiltrate AT after HF intake. However, it is unlikely that this underlies reduced NE-induced vasoconstriction in the present study, as nearly all the infiltrating immune cells around a vessel have been found in the perivascular fat/adventitia, not in medial areas (25), and our ex vivo experiments were performed on arteries cleared of surrounding AT. Furthermore, although the HF diet tended to decrease constrictor responses in WAT and BAT arteries, this effect was not universal. Indeed, no effect of the HF diet was found on KCl-mediated vasoconstriction in WAT or BAT arteries, nor was there an effect of the HF diet on ET-1 vasoconstriction in BAT. These findings argue against a generalized decrease in tone mediated by a vasodilator such as inducible nitric oxide synthase-derived nitric oxide after HF intake. Collectively, while the effects of a HF diet on BAT blood flow remain unclear, our results suggest that blood flow could be enhanced and that this may facilitate lipid uptake and, potentially, contribute to beiging of BAT. Conversely, beiging of BAT could also contribute to altered vasoconstrictor reactivity, resembling a more WAT-like phenotype.

To begin to pursue mechanisms underlying the altered vasoreactivity between adipose depots and after HF intake, we measured gene expression for selective agonist receptors in adipose arteries. Unfortunately, limited tissue availability precluded use of immunohistochemistry to evaluate protein expression and allowed us to measure gene expression for only a limited number of receptors. In general, agonist receptor gene expression did not change in parallel with the functional differences, suggesting that receptor density may not underlie the changes in vasoreactivity. However, protein analyses should confirm these negative findings, and the differences in arterial vasoreactivity should be used to inform future studies using pharmacodissection to describe the role of the specific agonist receptors.

Differences in the intrinsic mechanical properties between arteries from the adipose depots or alterations in response to a HF diet are other potential contributors to the changes in vasoconstrictor reactivity. To gain insight into this possibility, we evaluated the passive mechanical properties of the arteries, i.e., passive distension and circumferential stress and stretch. To our knowledge, this is the first comparison of these passive mechanical properties in arteries from WAT and BAT in NC-fed mice as well as the first study to examine the effect of a HF diet on mechanical properties of both WAT and BAT arteries. We found no differences in the morphological characteristics (lumen diameter, wall thickness, or wall-to-lumen ratio) or passive mechanical properties between adipose type or diet, suggesting that alterations in mechanical properties do not underlie depot or diet differences. Thus, further studies are required to elucidate the mechanisms underlying the differences in vasoconstrictor reactivity we observed between adipose depots and in response to the HF diet.

AT Markers

To our knowledge, this is the first study to demonstrate how a HF diet alters expression of WAT and BAT markers in both AT depots in B6D2F1 mice, thereby examining transdifferentiation in both adipose depots. We chose several traditional markers of WAT and BAT, and, by design, WAT markers (Pltp, Retn, Serpina3k, and Gsta3) were highly expressed in WAT relative to BAT, whereas expression of BAT markers (Ucp1, Cidea, and Elovl3) was higher in BAT than WAT in NC-fed mice.

WAT markers.

Expression of WAT-specific markers was unaffected in WAT of HF diet-fed mice, whereas expression of all WAT markers was greater in BAT of HF diet-fed mice. Although the mechanisms of the differential effect of the HF diet on AT require further investigation, this finding suggests an increase in white adipocyte differentiation in the BAT depot. To our knowledge, this is the first study examining the effect of a HF diet on expression of the WAT marker Serpina3k in either WAT or BAT. Furthermore, we found an increase in Retn expression in BAT of HF diet-fed mice, which is consistent with beiging of this tissue. In the present study, we did not find the previously reported increase in Retn expression in WAT after HF diet (28), but we did find that the HF diet was associated with greater expression of this gene in BAT. Although the mechanism for unaltered Retn expression in WAT is unclear, it may be due to the difference in diet duration (1 vs. 8 wk) between studies. Our results also demonstrate increased Pltp in BAT from HF diet-fed mice, which is consistent with beiging of this depot and supports previous findings that elevated activity of Pltp in the plasma of obese humans is reduced after weight loss (34). Further support for BAT beiging derives from our finding that the WAT marker Gsta3, a gene highly induced during adipocyte differentiation, tended (P = 0.06) to be upregulated in WAT and was highly induced in BAT from HF-fed mice, suggestive of enhanced white adipose differentiation in both adipose depots. Thus, it appears that a HF diet induces a shift in gene expression in BAT, favoring a more WAT-like, “beige” phenotype, which may have important consequences for systemic metabolism. This is a possibility requiring future investigation.

BAT markers.

Recently, it has become appreciated that there is metabolically active BAT in adult humans that may play an important role in the control of body weight (43). As such, understanding the impact of diet/obesity on BAT abundance and phenotype may have important clinical implications. Interestingly, several BAT markers are not exclusively expressed in BAT but are also present, although to a lesser extent, in normal WAT (54). Similar to previous studies that demonstrated increased expression of Ucp1 in BAT (36, 38) and either a decrease or no change in Ucp1 expression in WAT (14) after a HF diet, we found no change in Ucp1 expression after a HF diet in WAT. This finding is consistent with maintenance of white adipocytes, rather than a shift toward brown adipocytes, in this depot. Maintenance of the WAT phenotype would likely promote storage, rather than utilization, of excess dietary nutrients. The HF diet increased mRNA expression of another BAT marker, Elovl3, in BAT, similar to a previous report (39). However, we did not observe an effect of the HF diet on Cidea expression in either adipose depot, as has been previously reported (14). The disparate findings between the previous reports and the present study may be explained by animal strain differences, the age of the mice, or the composition and duration of the HF diet. It appears that a HF diet leads to a preferential increase in gene expression for WAT markers in BAT, consistent with beiging of BAT, as well as to increased expression of several BAT markers within BAT.

Limitations

Some limitations of this study should be noted. First, although not explored in the present study, an increase in body mass in the absence of increased caloric intake may result from decreased energy expenditure or a disruption of the normal feeding cycle in HF diet-fed mice (18, 20); this possibility should be better elucidated in future studies. Second, because of the small amount of tissue, we were unable to measure expression of the agonist receptors in the same arteries that were used in functional assessments. Rather, remaining arteries from the same adipose pads were excised and cleared of blood and surrounding tissues for RNA isolation and gene expression measures. As a result, differential expression related to location or branch order may mask our ability to detect relevant differences in receptor expression (32). Gene, but not protein, expression was measured for the same reason. Future studies should explore protein expression of the relevant receptors by Western blot analysis or immunohistochemistry. Third, we did not evaluate the secretomes of white and brown adipose depots. It has been previously reported that although WAT releases primarily leptin and adiponectin, BAT secretes a number of factors, including neuregulin 4, fibroblast growth factor 21, VEGF, bone morphogenetic proteins, nerve growth factor, and IL-6, as well as adiponectin (49), and these proteins could have profound effects on vascular reactivity. It is known, however, that BAT has a vasoprotective effect on small arteries (12) and that its removal can increase aortic stiffness (17), although the mechanisms/factors involved remain unclear. Removal of perivascular AT has also been shown to attenuate N-nitro-l-arginine methyl ester-induced reductions in lumen diameter, suggesting that the perivascular adipose is a source of endogenous nitric oxide (48), although the cell type responsible for its production remains elusive. While these are important questions for future investigations, we examined the reactivity of excised arteries cleared of surrounding tissues; thus, the role of the surrounding WAT/BAT and their secretome in modulating vasoconstrictor responses remains unclear.

Conclusions

In summary, the results of the present study demonstrate that, despite the potential for greater vasoconstriction in BAT than WAT arteries, as evidenced by greater KCl-mediated vasoconstriction, vasoconstriction in response to receptor-dependent agonists tended to be not different or lower in BAT than WAT. We further demonstrated that the HF diet reduced receptor-mediated vasoconstriction in arteries from WAT and BAT for most, but not all, agonists studied. The notable exception was ET-1, for which vasoconstriction was reduced in WAT but tended to increase in BAT arteries after the HF diet. Furthermore, neither alterations in receptor expression nor the mechanical properties of the arteries could explain the differences in reactivity. Finally, we provide evidence showing that, without altering tissue mass, a HF diet increases expression of genes related to both BAT and WAT within BAT. While the increase in the thermogenic BAT markers may be a compensatory mechanism to use excess dietary nutrients, increases in WAT markers provide evidence for beiging of BAT, which would promote lipid storage in this organ. Future studies should be focused on dissecting the mechanisms underlying differential changes in reactivity in these adipose arteries.

GRANTS

This work was supported in part by National Institute on Aging Grants R01-AG-048366, R01-AG-050238, and K02-AG-045339 and by Merit Review Award 1I01 BX-002151 from the United States Department of Veterans Affairs Biomedical Laboratory Research and Development Service.

DISCLAIMERS

The contents do not represent the views of the United States Department of Veterans Affairs, the National Institute on Aging, or the United States Government.

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

S.H., G.D.H., R.C.B., A.J.D., and L.A.L. performed experiments; S.H. and L.A.L. analyzed data; S.H., A.J.D., and L.A.L. interpreted results of experiments; S.H. prepared figures; S.H. drafted manuscript; S.H., G.D.H., A.J.D., and L.A.L. edited and revised manuscript; S.H., G.D.H., R.C.B., A.J.D., and L.A.L. approved final version of manuscript.