Neural Programmatic Role of Leptin, TNFα, Melanocortin, and Glutamate in Blood Pressure Regulation vs Obesity-Related Hypertension in Male C57BL/6 Mice

Abstract

Continuous nutritional surplus sets the stage for hypertension development. Whereas moderate dietary obesity in mice is normotensive, the homeostatic balance is disrupted concurrent with an increased risk of hypertension. However, it remains unclear how the obesity-associated prehypertensive state is converted into overt hypertension. Here, using mice with high-fat-diet (HFD)–induced moderate obesity vs control diet (CD)–fed lean mice, we comparatively studied the effects of central leptin and tumor necrosis factor-α (TNFα) as well as the involvement of the neuropeptide melanocortin pathway vs the neurotransmitter glutamate pathway. Compared with CD-fed lean mice, the pressor effect of central excess leptin and TNFα, but not melanocortin, was sensitized in HFD-fed mice. The pressor effect of central leptin in HFD-fed mice was strongly suppressed by glutamatergic inhibition but not by melanocortinergic inhibition. The pressor effect of central TNFα was substantially reversed by melanocortinergic inhibition in HFD-fed mice but barely in CD-fed mice. Regardless of diet, the hypertensive effects of central TNFα and melanocortin were both partially reversed by glutamatergic suppression. Hence, neural control of blood pressure is mediated by a signaling network between leptin, TNFα, melanocortin, and glutamate and changes in dynamics due to central excess leptin and TNFα mediate the switch from normal physiology to obesity-related hypertension.

The central nervous system clearly plays an important role in regulation of blood pressure (BP) homeostasis (1,,,,–6). Among various responsible brain regions, the hypothalamus has received research attention. For example, several hormonal factors, such as leptin, melanocortin, angiotensin, and tumor necrosis factor-α (TNFα), have been reported for a role in hypothalamic BP regulation (7,,,,,,–14). Because the hypothalamus is essential for metabolic regulation as well as the development of metabolic diseases such as obesity, the pressor actions of these hypothalamic factors are conceivably intertwined with metabolic and nutritional statuses. Although overnutrition-induced obesity and obesity-related hypertension are increasing and are worldwide risk factors to human health (15, 16), several studies have demonstrated the elevated serum leptin in animals after high-fat diet (HFD) feeding or in obese human subjects (17, 18). Moreover, a close association between BP and serum leptin levels has been established (19). In parallel, chronic inflammation is commonly associated with both obesity and hypertension (13, 20) and, recently, the central inflammatory mechanisms underlying the pathogenesis of obesity-related hypertension have become a key focus of investigations (13). Leptin infusion in the hypothalamus and excess inflammatory factors, such as TNFα, have been shown to be involved in the development of obesity-related hypertension (13, 21). The neuronal basis for hypothalamic regulation of BP is linked to the melanocortin system, which emerges as a crucial pathway in the development of metabolic disease–associated hypertension and other cardiovascular disorders (22). Experimental evidence in normal rats showed that α-melanocyte-stimulating hormone (α-MSH), a key neuropeptide in the melanocortin system, induced hypertensive effects, which could be partially reversed by SHU9119, an inhibitor of melanocortin receptors MC3/4R (9). Glutamatergic neurons are also an important regulator of BP. For example, intracerebral ventricular injection of glutamate can excite hypothalamic neurons and elicit an increase in arterial BP, heart rate, and the sympathetic activity. Blockade of hypothalamic glutamate receptors by kynurenate acid (KYN) can partially reverse the hypertensive effect induced by leptin injection in normal control diet (CD)–fed lean rats (9). However, despite our understanding thus far, there is still a lack of knowledge regarding how these various factors work interactively to regulate BP and especially how the system might evolve to switch from a homeostatic role to a pathogenic role stimulating obesity-related hypertension.

Methods

Animals

Male C57BL/6 mice with 3-month HFD feeding vs CD feeding were obtained from Jackson Laboratory. Upon completion of designed feeding regimen, mice (4 to 5 months old) were acclimated for at least 5 days before surgery with free access to food and water on a 12-hour/12-hour light/dark cycle with controlled temperature (23°C ± 2°C) and humidity (50% ± 5%). Mice continued to have ad libitum access to HFD (60% kcal fat; Research Diets, Inc., New Brunswick, NJ) vs standard normal CD (11% kcal fat; LabDiet, St. Louis, MO) throughout experimentation. All of the animal procedures were approved by the Institutional Animal Care and Use Committee at the Albert Einstein College of Medicine.

Telemetric probe implantation and cannulation

We measured arterial BP on conscious, freely moving mice using a radiotelemetry monitoring system (Data Sciences International, New Brighton, MN). The surgical procedure for radiotelemetric probe (model TA11PA-C10; Data Sciences International) implantation and third-ventricle cannulation were performed as described previously (14, 23). Briefly, mice were anesthetized, and a midline skin incision was made from the tip of the mandible to the sternal notch. The left common carotid artery was isolated; whereas the proximal end of the vessel was ligated, the distal end was occluded with a microclip. Through an incision made near the proximal end, a pressure transmission catheter was guided into the artery, and the transmitter body was placed into the subcutaneous pocket. After probe implantation, the neck incisions were closed using 5-0 sutures (Ethicon Inc., Somerville, NJ). We implanted a guide cannula (26-gauge; Plastics One, Inc., Roanoke, VA) into the third ventricle of anesthetized mice at the midline coordinates 1.82 mm posterior to the bregma and 5.0 mm below the skull. Mice were allowed to fully recover from the surgery.

Pharmacological injection

After a 1-week postimplantation recovery period, mice were injected with chemicals, including leptin (2.5 μg; R&D Systems, Minneapolis, MN), TNFα (10 pg; Sigma-Aldrich, St. Louis, MO), melanocortin 3/4 receptors antagonist SHU9119 (1 nM; Tocris Bioscience, Bristol, UK), ionotropic glutamate receptor inhibitor kynurenic acid (20 mM; Tocris Bioscience), and α-MSH (10 μg; Sigma-Aldrich). Chemicals were dissolved in 0.5 µL of artificial cerebrospinal fluid (Harvard Apparatus, Holliston, MA) and injected into the hypothalamic third ventricle of mice through the preimplanted cannula over a 5-minute period using a 33-gauge internal injector (Plastics One, Inc.) that was connected to a 5-µL Hamilton syringe.

BP recording and data acquisition

BP was recorded in the daytime phase (9:00 to 15:00) to avoid the influence of endogenous TNFα and leptin circadian patterns on BP fluctuations (14, 24). BP and heart rate were relatively stable and less volatile during this period (25, 26). Data were sampled continuously with a sampling rate of 1,000 Hz with 1-minute segment duration.

Body composition

Body composition was assessed by quantitative magnetic resonance imaging for analysis of fat and lean mass calibrated by EchoMRI^TM Whole Body Composition Analyzer (frequency: 2284.9 Hz; calibration coefficient: 0.92; Echo Medical System, Houston, TX).

Tissue harvest and enzyme-linked immunosorbent assay (biochemical assays)

CD- and HFD-fed mice were killed between 10:00 and 11:00; serum and hypothalamus were collected as described previously (23). TNF concentrations in the serum and hypothalamus, and leptin levels in the serum, were determined with mouse TNFα and leptin enzyme-linked immunosorbent assay kits according to the manufacturer’s instructions (Life Technologies, Carlsbad, CA).

Statistical analyses

Data are presented as means ± standard error of the mean. A two-tailed, unpaired t test was used for studies that involved only two groups. Analysis of variance and appropriate post hoc analyses were used for experiments comprising more than two groups. P < 0.05 was considered statistically significant.

Results

Moderate obesity is associated with elevated leptin and TNFα but normal BP

In this study, we used male adult C57BL/6 mice with 3-month HFD feeding as an animal model to mimic a prehypertensive state with moderate dietary obesity (14, 27). These mice were compared with age-matched, CD-fed lean mice. Mice placed on a 3-month regimen of HFD feeding developed moderate obesity as shown by increased body weight and fat mass (Table 1). Telemetric recording revealed that these HFD-fed mice showed comparable BP to normal CD-fed animals. Average systolic BP was around 120 mm Hg, and average diastolic BP was around 100 mm Hg (Table 1). Thus, mice fed with 3-month HFD were moderately obese but normotensive. We analyzed serum and hypothalamus samples from a group of these mice; as expected, serum leptin levels remarkably increased in HFD-fed mice (Table 1), which is consistent with formation of leptin resistance (17, 28, 29). Serum TNFα was undetectable in CD-fed mice but reached 4.6 pg/mL in HFD-fed mice, and similarly, the hypothalamic TNFα level was higher in HFD-fed mice than CD-fed mice (Table 1). Thus, despite the fact that these mice are normotensive, the risk factors for hypertension development are constantly accumulating, including the circulating metabolic signals, a proinflammatory microenvironment, and neuroendocrine/neural pathways that control BP (20, 30).

Table 1.

Baseline Profiles in Normal CD-Fed and HFD-Fed Mice

	CD (n = 5)	HFD (n = 5)
Body weight, g	25.3 ± 0.4	35.3 ± 1.5a
Fat mass, g	2.8 ± 0.1	11.5 ± 1.4a
Lean mass, g	21.5 ± 0.5	22.8 ± 0.9
SBP, mm Hg	117.5 ± 7.6	121.9 ± 7.9
DBP, mm Hg	98.5 ± 5.4	97.5 ± 5.4
MBP, mm Hg	105.4 ± 6.7	107.0 ± 6.6
HR, beats/min	523.1 ± 24.2	541.3 ± 34.7
Serum leptin, ng/mL	1.54 ± 0.46	13.3 ± 1.0a
Serum TNFα, pg/mL	n.d.	4.6 ± 2.8
Hypothalamus TNFα, pg/mg	0.45 ± 0.02	0.55 ± 0.02b

Statistical analyses: mean ± standard error of the mean.

Abbreviations: DBP, diastolic blood pressure; HR, heart rate; MBP, mean blood pressure; n.d., not detectable; SBP, systolic blood pressure.

^a

P < 0.01 (two-tailed Student t test).

^b

P < 0.05 (two-tailed Student t test).

Leptin-induced BP rise in obesity is not reversed by melanocortinergic inhibition

HFD-fed mice were implanted with a telemetric BP probe in the carotid artery and an injection cannula in the third ventricle (23, 30). Following the postimplantation recovery period, we injected leptin (2.5 μg) into the hypothalamic third ventricle but did not notice a BP-raising effect from a single injection. It was previously found that the hypertensive effect of leptin is not obvious following a single injection at this dose (14). However, microinjection of leptin into HFD-fed mice for 2 consecutive days resulted in increased BP and heart rate (Fig. 1). BP increased roughly from 40 to 160 minutes postinjection [Fig. 1(e–h)]. Consistent with our previous observation (14), leptin-induced hypertension in obese mice displayed an early and late phase of increase BP (Fig. 1). To understand the neural pathways that might underlie leptin’s pressor effect in HFD-fed mice, we examined the melanocortin pathway, because central leptin is known to activate propiomelanocortin neurons, leading to the release of α-MSH, which then acts on MC3/4R of secondary neurons to increase the sympathetic outflow (31,,–34). It is well documented that central administration of either α-MSH or an MC3/4R agonist increases the sympathetic outflow (9, 15). To address whether an excess of central leptin could activate MC3/4R signaling leading to hypertension in obesity, we used an intrahypothalamic injection of an MC3/4R antagonist, SHU9119 (1 nM). Surprisingly, SHU9119 did not obviously inhibit the pressor effects of central excess leptin (Fig. 1). Hence, we speculate that, as a result of hypothalamic microenvironmental change from HFD feeding, additional neural pathway(s) might participate in the mechanism behind central leptin-induced hypertension in obesity.

Figure 1.

Effect of melanocortinergic inhibition on leptin-induced BP rise in obesity. Male C57BL/6 mice fed on a HFD for 3 months were implanted with a BP radio transmitter in the carotid artery and an injection cannula in the hypothalamic third ventricle. After 1 week of postsurgery recovery, mice received daily injection of leptin through the cannula for 2 consecutive days. On the second day of injection, HFD-fed mice received an intra–third-ventricle injection of SHU9119 vs vehicle 30 minutes prior to leptin administration. Curves on the left present the minute-by-minute average levels of (a) systolic, (b) diastolic, and (c) mean BP and (d) heart rate over a 4-hour postinjection period. Bar graphs on the right present average levels of (e) systolic, (f) diastolic, and (g) mean BP and (h) heart rate during two phases (35 to 65 minutes and 90 to 150 minutes after injection) of leptin-induced hypertension in HFD-fed mice. Phase 1 (35 to 65 minutes) is outlined by a gray dotted line (a–d), and phase 2 (90 to 150 minutes) is outlined by a red dotted line (a–d). Error bars reflect mean ± standard error of the mean. *P < 0.05, **P < 0.01; n = 5 mice per group. aCSF, artificial cerebrospinal fluid; bpm, beats per minute; DBP, diastolic blood pressure; HR, heart rate; Inj & Rec, injection and recovery; Lep, leptin; MBP, mean blood pressure; n.s., nonsignificant; SBP, systolic blood pressure; SHU, SHU9119.

Reversal of leptin-induced BP rise in obesity by glutamatergic inhibition

According to our results in Fig. 1, the leptin-melanocortin connection seemed to have a negligible or modest contribution to the pathogenesis of obesity-related hypertension. To explore alternative mediators, we examined the glutamatergic pathway, considering that the physiological actions of excitatory neurotransmitter glutamate have often been linked to BP increase (35, 36). To be experimentally comparable, we used the same model as shown in Fig. 1 but performed a single intrahypothalamic third-ventricle injection of the glutamatergic inhibitor KYN (20 mM) 30 minutes prior to the second-day leptin injection. We found that KYN administration significantly reversed the pressor effects of leptin in these mice (Fig. 2). The antihypertensive effect of KYN was most notable between 35 and 65 minutes and between 105 and 165 minutes after leptin injection, which, respectively, corresponded to the early and late phase of leptin-induced hypertension in these mice (Fig. 2). These findings suggested that in obesity, excess central leptin enhanced, directly or indirectly, glutamatergic pathway activation, which increased the BP. Altogether, inhibition of the ionotropic glutamate receptor in the obesity condition greatly reversed leptin-induced hypertension, as opposed to melanocortinergic inhibition, which had little effect. This highlighted the importance of the glutamatergic pathway in the development of hypertension with excess central leptin.

Figure 2.

Effect of glutamatergic inhibition on leptin-induced BP rise in obesity. Male C57BL/6 mice fed on a HFD for 3 months followed the same surgical procedure as described in Fig. 1. After 1-week postsurgery recovery, mice received daily intra–third-ventricle injection of leptin for 2 consecutive days. In the second day, mice received an intra–third-ventricle injection of KYN vs vehicle 30 minutes prior to leptin administration. Curves on the left present the minute-by-minute average levels of (a) systolic, (b) diastolic, and (c) mean BP and (d) heart rate over a 4-hour postinjection period. Bar graphs on the right present average levels of (e) systolic, (f) diastolic, and (g) mean BP and (h) heart rate during early and late phases (35 to 65 minutes and 105 to 165 minutes postinjection) of leptin-induced hypertension in HFD-fed mice. Phase 1 (35 to 65 minutes) is outlined by a gray dotted line (a–d), whereas phase 2 (105 to 165 minutes) is outlined by a red dotted line (a–d). Error bars reflect mean ± standard error of the mean. *P < 0.05, **P < 0.01; n = 5 mice per group. aCSF, artificial cerebrospinal fluid; bpm, beats per minute; DBP, diastolic blood pressure; HR, heart rate; Inj & Rec, injection and recovery; Lep, leptin; MBP, mean blood pressure; n.s., nonsignificant; SBP, systolic blood pressure.

Lack of hypertensive development with central leptin administration in lean mice

In conjunction with studies of HFD-fed mice as presented in Figs. 1 and 2, we performed the same experiments but used age-matched, CD-fed lean mice (Table 1). In contrast to the hypertensive effect of central excess leptin in Figs. 1 and 2, hypothalamic third-ventricle injection of leptin using the same procedure did not cause any evident BP increase in regular CD-fed lean mice (Fig. 3). On day 2 we also injected SHU9119 (1 nM) into the hypothalamic third ventricle 30 minutes prior to leptin injection, similar to Fig. 1, but did not observe any obvious BP reduction in these mice (Fig. 3). Separately, hypothalamic third-ventricle injection of KYN (20 mM) in these mice also did not lower their BP (Fig. 3). In addition, it is worth mentioning that neither of these two pharmacological treatments decreased the BP of vehicle-injected control mice (Fig. 3), indicating that the experimental level of melanocortinergic or glutamatergic inhibition in this study did not affect normal BP in physiologically normal animals.

Figure 3.

Lack of hypertensive mechanism for central leptin in CD-fed lean mice. Male CD-fed C57BL/6 mice (with matched ages with HFD feeding) received the same surgical procedure as described in Figs. 1 and 2. After 1-week postsurgery recovery, mice received daily intra–third-ventricle injection of leptin for 2 days. On the second day of injection, mice were injected with SHU9119, KYN, or vehicle in the hypothalamic third ventricle 30 minutes prior to leptin injection. Curves on the left present the minute-by-minute average levels of (a) systolic, (b) diastolic, and (c) mean BP and (d) heart rate over a 4-hour postinjection period. Bar graphs on the right present average levels of (e) systolic, (f) diastolic, and (g) mean BP and (h) heart rate during early and late phases (35 to 65 minutes and 105 and 165 minutes postinjection) of leptin-induced hypertension in HFD-fed mice. Phase 1 (35 to 65 minutes) is outlined by a gray dotted line (a–d), and phase 2 (105 to 165 minutes) is outlined by a red dotted line (a–d). Error bars reflect mean ± standard error of the mean. n = 5 mice per group. aCSF, artificial cerebrospinal fluid; bpm, beats per minute; DBP, diastolic blood pressure; HR, heart rate; Inj & Rec, injection and recovery; Lep, leptin; MBP, mean blood pressure; SBP, systolic blood pressure; SHU, SHU9119.

Melanocortinergic inhibition partially reverses TNFα-induced BP rise in obesity

We have previously reported that obese mice are more prone to hypothalamic inflammation (30). Additionally, central TNFα injection increased BP through increased sympathetic outflow (30). With this background, we comparatively studied how central excess TNFα might differentially affect the BP of mice in normal health vs an obesity condition. We injected 10 pg of TNFα into the hypothalamic third ventricle of CD-fed vs HFD-fed mice using a protocol we established previously (14, 30). As shown in Fig. 4, a single TNFα injection clearly increased BP and heart rate in both CD-fed and HFD-fed mice, with a much stronger hypertensive effect in HFD-fed mice than CD-fed mice. Thus, an obesity-associated proinflammatory state potentiates the hypertensive action of central excess TNFα. Subsequently, as similarly performed to study leptin’s pressor effect, we tested whether melanocortinergic inhibition could interfere with the hypertensive action of TNFα. To do so, mice were centrally administered SHU9119 and, 30 minutes later, TNFα. In CD-fed mice, SHU9119 treatment had little inhibitory effect on TNFα-induced hypertension [Fig. 4(a–e)]. Conversely, SHU9119 treatment in HFD-fed mice showed a strong, although incomplete, potential to prevent TNFα-induced increased BP and heart rate [Fig. 4(f–j)]. Thus, differing from the mechanism of leptin-induced hypertension, central excess TNFα in obesity significantly used the melanocortinergic pathway to induce hypertension and related cardiovascular disorders.

Figure 4.

Effect of melanocortinergic inhibition on TNFα-induced BP rise in obesity. Male C57BL/6 mice after 3-month HFD vs CD feeding received the same surgical procedure as described in Figs. 1–3. After surgery and recovery, mice received an intra–third-ventricle injection of SHU9119 vs vehicle 30 minutes prior to TNFα injection. Curves present the minute-by-minute average levels of (a, f) systolic, (b, g) diastolic, and (c, h) mean BP and (d, i) heart rate in (a–d) CD-fed vs (f–i) HFD-fed mice over a 4-hour postinjection period. Bars show the average levels of BP and heart rate in (e) CD-fed vs (j) HFD-fed mice during the 50 to 100 minutes postinjection outlined by a dotted line. Error bars reflect mean ± standard error of the mean. *P < 0.05, **P < 0.01; n = 5 mice per group. aCSF, artificial cerebrospinal fluid; bpm, beats per minute; DBP, diastolic blood pressure; HR, heart rate; Inj & Rec, injection and recovery; MBP, mean blood pressure; n.s., nonsignificant; SBP, systolic blood pressure; SHU, SHU9119; TNF, TNFα.

Glutamatergic inhibition strongly reverses TNFα-induced BP rise in obesity

In addition to melanocortin inhibition, we studied whether glutamatergic inhibition could affect TNFα-induced hypertension in HFD-fed vs CD-fed mice. The experimental setup and procedures were the same as presented in Fig. 2 except that we used a single injection of TNFα vs the vehicle. In CD-fed mice, central blockade of glutamate using KYN was found to partially reverse TNFα-induced hypertension [Fig. 5(a–e)]. This antihypertensive effect with glutamatergic inhibition contrasted with the absence of such an effect using the melanocortinergic inhibitor SHU9119 [Fig. 4(a–e)]. The ability of central KYN injection to inhibit TNFα-induced hypertension was similarly seen in HFD-fed mice [Fig. 5(f–j)]. The ability of central KYN injection to lower the BP of HFD-fed mice after TNFα injection was in agreement with that seen with SHU9119 [Fig. 4(f–j)]. Notably, the BP-lowering effect of KYN seemed to lag in its time course, compared with the time course of SHU9119, potentially suggesting that TNFα-induced glutamatergic activation might lie downstream of melanocortinergic activation. Also, in comparing the obesity-associated hypertensive action of TNFα and leptin, whereas the melanocortin pathway was differentially involved, the glutamatergic pathway was similarly effective at suppressing the hypertensive effect of both factors in obesity.

Figure 5.

Effect of glutamatergic inhibition on TNFα-induced BP rise in obesity. Male C57BL/6 mice after 3-month HFD vs CD feeding received the same surgical procedure as described in Figs. 1–3. After surgery and recovery, mice received an intra–third-ventricle injection of KYN vs vehicle 30 minutes prior to TNFα injection. Curves present the minute-by-minute average levels of (a, f) systolic, (b, g) diastolic, and (c, h) mean BP and (d, i) heart rate in (a–d) CD-fed vs (f–i) HFD-fed mice over a 4-hour postinjection period. Bars show the average levels of BP and heart rate in (e) CD-fed (50 to 100 minutes postinjection) vs (j) HFD-fed (BP, 65 to 115 minutes postinjection; heart rate, 40 to 90 minutes postinjection) mice outlined by a dotted line (a–d, f–i). Error bars reflect mean ± standard error of the mean. *P < 0.05, **P < 0.01; n = 5 mice per group. aCSF, artificial cerebrospinal fluid; bpm, beats per minute; DBP, diastolic blood pressure; HR, heart rate; Inj & Rec, injection and recovery; MBP, mean blood pressure; n.s., nonsignificant; SBP, systolic blood pressure; TNF, TNFα.

Glutamatergic inhibition partially reverses α-MSH-induced BP rise in obesity

As presented earlier, the glutamatergic pathway could potentially lie downstream of melanocortinergic activation to increase BP in both CD-fed and HFD-fed conditions. To test this possibility, we performed experiments in which we injected α-MSH (10 μg) into the hypothalamic third ventricle of CD-fed and HFD-fed mice. About 30 minutes after injection with α-MSH, both CD-fed and HFD-fed mice displayed increases in BP, with mean BP values reaching approximately 120 to 130 mm Hg and plateauing for approximately 90 minutes (Fig. 6). Interestingly, we noted that this hypertensive effect was not stronger in HFD-fed mice compared with CD-fed mice (Fig. 6), suggesting that the melanocortin pathway is important for BP regulation in normal physiology, whereas its contribution to the development of hypertension in the obesity condition requires additional pathogenic factors. Of importance, we examined whether the BP increase in these models could be inhibited by glutamatergic blockade. As shown in Fig. 6, although heart rate did not seem to be decreased, KYN microinjection greatly decreased the hypertensive effect of α-MSH, regardless of diet. Putting together all the results, we deduced a model (Fig. 7): Melanocortinergic activation by central TNFα in obese mice can lead to glutamatergic activation to increase BP. Comparatively, the hypertensive action of leptin in obese mice appears to mostly use glutamatergic signaling. Despite the divergent mechanisms, glutamatergic upregulation can be seen in the hypertensive effects of both TNFα and leptin in obesity; therefore, inhibition of glutamatergic signaling bears an important potential to combat obesity-related hypertension.

Figure 6.

Effect of glutamatergic inhibition on α-MSH-induced BP rise in obesity. Male C57BL/6 mice after 3-month HFD vs CD feeding received the same surgical procedure as described in Figs. 1–3. After surgery and recovery, mice received an intra–third-ventricle injection of KYN vs vehicle 30 minutes prior to α-MSH injection. Curves present the minute-by-minute average levels of (a, f) systolic, (b, g) diastolic, and (c, h) mean BP and (d, i) heart rate in (a–d) CD-fed vs (f–i) HFD-fed mice over a 4-hour postinjection period. Bars show the average levels of BP and heart rate in (e) CD-fed (85 to 155 minutes after injection) vs (j) HFD-fed (40 to 110 minutes after injection) mice outlined by a dotted line (a–d, f–i). Error bars reflect mean ± standard error of the mean. *P < 0.05, **P < 0.01; n = 5 mice per group. aCSF, artificial cerebrospinal fluid; bpm, beats per minute; DBP, diastolic blood pressure; HR, heart rate; Inj & Rec, injection and recovery; MBP, mean blood pressure; MSH, α-MSH; n.s., nonsignificant; SBP, systolic blood pressure.

Figure 7.

Neural program in BP regulation vs obesity-related hypertension. Central control of BP in normal physiology involves neural actions of leptin and TNFα; whereas previous work revealed how the connection between leptin and TNFα affects BP, the current study further elucidates that leptin and TNFα also use the melanocortinergic pathway and glutamatergic pathway, respectively, to mediate the physiological control of BP. Crosstalk between these two processes at the level of the downstream neuronal circuitry was identified. In the obesity condition, increased leptin release activates glutamatergic neurons, which contributes to hypertension mainly independently of melanocortinergic neurons, whereas an obesity-associated increase in TNFα (partially due to high leptin level) activates melanocortinergic and glutamatergic neurons, both sequentially and in parallel, to mediate the induction of hypertension. Dotted lines indicate the weakened pathway in obesity compared with normal physiology. Bold lines indicate the pathways that become predominantly important steps in this complex mechanism in obesity. In general, glutamatergic activation is poised at the crossroad of translating obesity-related signals to hypertension and thus represents a critical target for combating obesity-related hypertension.

Discussion

The hypothalamus is a key neuroendocrine organ known to regulate energy homeostasis via specific and distinct neural pathways and neuroendocrine hormones that regulate energy balance and nutrient homeostasis (37). Moreover, it is a key site in the brain that integrates central and peripheral inputs to ultimately impact BP in multiple disease states (30, 38). An increase in nutritional intake can elevate the levels of several hypothalamic signals, such as leptin and TNFα, which act in many regions to initiate their sympatho-excitatory effects (17, 18, 28, 29, 33, 39). Our recent study has provided evidence showing that hypothalamic leptin and TNFα participate in the circadian, physiological control of BP fluctuation. Leptin employs TNFα as a mediator to induce an increase in BP during diurnal cycles in normal physiology (14). Using this background, this work was designed to further explore the downstream neural networks responsible for the BP-raising action of TNFα in a normal, lean condition. We found that the BP-raising effect of TNFα in normal, lean mice depends on the glutamatergic pathway rather than the melanocortinergic pathway. This finding is in agreement with the literature showing that TNFα often utilizes the glutamatergic pathway in many other aspects of neurologic functions (40,–42). In addition, we studied the relationship between the melanocortinergic pathway and the glutamatergic pathway and found that glutamatergic inhibition greatly, although partially, reversed the hypertensive action of melanocortinergic activation. Hence, as summarized in Fig. 7, our work suggests that leptin and TNFα can use the melanocortinergic pathway and glutamatergic pathway, respectively, to mediate the physiological control of BP, and crosstalk between these two processes occurs through downstream neuronal circuitry. Because the BP-raising action of leptin in the normal condition requires repeated injections as well, as in the context of our previous finding (14), we postulate that the physiological function of the leptin-melanocortinergic pathway is to maintain BP levels, including the diurnal rhythm of BP. However, the TNFα-glutamatergic pathway functions to elevate BP in response to physiological or physical demand.

In comparison with normal physiology, we used a mouse model for moderate obesity in a prehypertensive state to investigate whether these four factors might integrate and therefore contribute to hypertension development in obesity. The homeostatic balance between the peripheral and central nervous systems in the HFD feeding condition undergoes repeated disruptions and repairs, gradually leading to the induction of obesity and leptin resistance. This is associated with greatly enhanced leptin release and thus elevated leptin concentrations in the blood as well as the cerebrospinal fluid. This obesity condition is also closely associated with a state of hypothalamic inflammation because chronic overnutrition induces inflammation-like changes in the hypothalamus (23, 43,,,,,–49). It has been reported that inhibition of TNFα in the hypothalamus delays hypertensive development in spontaneously hypertensive rats, suggesting a crucial role for TNFα in the pathogenesis of hypertension (49). Our previous work has demonstrated that central excess TNFα is a cause of hypertension in obesity (14, 30). In this study, we used mice with moderate obesity induced by 3-month HFD feeding, a pathophysiological state reflecting an early-stage inflammatory situation in which the basal leptin and TNFα levels have increased in these mice while their BP and heart rate are still normal. Our results show that compared with the effects in CD feeding, HFD feeding sensitizes the pressor effects of central excess leptin and TNFα. Importantly, compared with the normal physiology in which the leptin-melanocortinergic pathway and the TNFα-glutamatergic pathway are separately activated to regulate BP, we found that these two pathways are cross-linked in the obesity condition, and thus their effects are synergized, which significantly contributes to pathological hypertension (Fig. 7). In the condition of obesity with leptin resistance, the hypertensive effect of hyperleptinemia might not be directly mediated by the melanocortinergic pathway but alternatively through the glutamatergic pathway as well as directly through the TNFα-melanocortinergic pathway. In addition, excess TNFα from other obesity-associated sources also employs the glutamatergic pathway to increase the extent of hypertension. Overall, the glutamatergic activation is a convergence point in linking leptin, TNFα, and melanocortin signaling to drive obesity-related hypertension and could represent a target in developing therapeutic strategies against obesity-related hypertension.

Abbreviations:

 
• BP

  blood pressure

 
• CD

  control diet

 
• HFD

  high-fat diet

 
• KYN

  kynurenate acid

 
• α-MSH

  α-melanocyte-stimulating hormone

 
• TNFα

  tumor necrosis factor-α.

Acknowledgments

We thank Dr. Cai’s laboratory members for technical assistance.

This work was supported by National Institutes of Health Grants R01 DK078750, R01 AG031774, R01 HL113180, and R01 DK099136 (to D.C.).

Disclosure Summary: The authors have nothing to disclose.

Author notes

Address all correspondence and requests for reprints to: Dongsheng Cai, MD, PhD, Department of Molecular Pharmacology, Albert Einstein College of Medicine, 1300 Morris Park Avenue, Bronx, New York 10461. E-mail: [email protected].