Vascular heme oxygenase (HO) metabolizes heme to form carbon monoxide (CO). Increased heme-derived CO inhibits nitric oxide synthase and can contribute to hypertension via endothelial dysfunction in Dahl salt-sensitive rats. Obese Zucker rats (ZR) are models of metabolic syndrome. This study tests the hypothesis that endogenous CO formation is increased and contributes to hypertension and endothelial dysfunction in obese ZR. Awake obese ZR showed increased respiratory CO excretion, which was lowered by HO inhibitor administration [zinc deuteroporphyrin 2,4-bis glycol (ZnDPBG) 25 μmol·kg−1·24 h−1 ip]. In awake obese ZR, chronically instrumented with femoral arterial catheters, blood pressure was elevated but was decreased by the HO inhibitor ZnDPBG. Body weight, blood glucose, glycated hemoglobin, plasma insulin, total and LDL cholesterol, oxidized LDL, and triglyceride levels were elevated in obese ZR, and, except for LDL cholesterol, were unchanged by HO inhibition. Total HO-1 protein levels were not different between lean and obese ZR aortas. In vitro experiments used isolated skeletal muscle arterioles with constant pressure and no flow, or constant midpoint, but altered endpoint pressures to establish graded levels of luminal flow. In obese ZR arterioles, responses to ACh and flow were attenuated. Acute in vitro pretreatment with an HO inhibitor, chromium mesoporphyrin, enhanced ACh and flow-induced dilation and abolished the differences between groups. Furthermore, exogenous CO prevented the restoration of flow-induced dilation by the HO inhibitor in obese ZR arterioles. These results suggest that HO-derived CO production is increased and promotes hypertension and arteriolar endothelial dysfunction in obese ZR with metabolic syndrome independent of affecting metabolic parameters.
- heme oxygenase
- blood pressure
- vascular tone regulation
metabolic syndrome is characterized by upper body obesity, insulin resistance, hyperlipidemia, and hypertension, and it represents a major and growing health problem in the United States (11). Endothelial dysfunction due to impaired nitric oxide function, is a key feature promoting microvascular disease in resistance vessels, which contributes to hypertension during obesity (43) and metabolic syndrome. The obese Zucker rat (ZR) is a well established genetic model of obesity (58). Because of a nonfunctional leptin receptor gene (19) and the consequent hyperphagia, these animals develop obesity and metabolic syndrome, including insulin resistance (29), hyperlipidemia (29), and hypertension (13). Previous studies have shown impaired endothelium-derived nitric oxide-dependent vasodilation in various vascular beds of obese ZR (12, 13, 33, 41), including skeletal muscle resistance vessels (12, 13).
Metabolic syndrome represents a major cardiovascular risk factor, which is further enhanced by the increased chance of type 2 diabetes development (11). It is known that patients with type 2 diabetes show increased respiratory carbon monoxide (CO) levels (39). Although elevated blood glucose levels in obese ZR are not as prominent as some other diabetes models, it is possible that the elevated glucose, in conjunction with other obesity risk factors, could promote metabolic and developmental changes to promote CO production.
The primary endogenous source of CO formation is via enzymatic breakdown of heme by heme oxygenase (HO) (47). Numerous tissues (35), including vascular endothelial and smooth muscle cells express, HO (8, 16, 24). Two major enzymatically active isoforms of HO (35) are the inducible HO-1 and the constitutive HO-2. Pathological conditions (35), such as ANG II-induced (20), Dahl/Rapp salt-sensitive (21, 48), or DOCA-salt hypertension (22) can increase vascular HO-1 expression. Although heme-derived CO can relax vascular smooth muscle (10, 15, 27) and protect against some aspects of apoptosis (9), it also interferes with the vasodilator effects of the nitric oxide system (37, 49, 56) and promotes endothelium-dependent vasoconstriction (23, 42).
Because patients with elevated glucose can present with elevated levels of CO production (39) and because CO can inhibit nitric oxide synthase (37, 49, 56) and promote hypertension, we hypothesized that heme-derived CO formation is increased and contributes to hypertension and arteriolar endothelial dysfunction in obese ZR. To test this hypothesis, we used awake lean and obese ZR rats and examined the effects of an inhibitor of HO on blood pressure and CO production. To examine endothelial function, we also conducted experiments using skeletal muscle arterioles taken from age-matched lean and obese ZR, and we examined the responses to an endothelium-dependent vasodilator and increases in luminal flow while in the presence or absence of an inhibitor of HO.
MATERIALS AND METHODS
Zinc deuteroporphyrin 2, 4-bis ethylene glycol (ZnDPBG) and chromium mesoporphyrin (CrMP) were purchased from Frontier Scientific (Logan, UT); these metalloporphyrins were used for this study as they are potent (51) and selective (3) inhibitors of HO activity. Although solubility and photostability (51) make CrMP a preferred choice for in vitro studies, ZnDPBG is ideal for in vivo experiments because of its quick distribution (52). Isoflurane was obtained from Abbott Laboratories (North Chicago, IL). Thiobutabarbital sodium (Inactin), and ACh were obtained from Sigma (St. Louis, MO). All other chemicals were obtained from Fisher Scientific (Houston, TX). ACh (10 mmol/l) stock solution was prepared in saline and diluted in modified Krebs buffer immediately before use. Inactin was prepared fresh daily in saline (50 mg/ml). CrMP stock solution (15 mmol/l) was prepared in 50 mmol/l Na2CO3 solution and diluted in modified Krebs buffer (15 μmol/l) immediately before use. ZnDPBG solution (15 mmol/l) was prepared in 50 mmol/l Na2CO3 solution for in vivo administration. The composition of modified Krebs buffer was (mmol/l): 118.5 NaCl; 4.7 KCl; 1.4 CaCl2; 1.2 KH2PO4; 1.1 MgSO4; 25.0 NaHCO3; and 11.1 dextrose.
Male lean (Fa/Fa or Fa/fa, n = 48) and obese (fa/fa, n = 61) Zucker rats (ZR) (Harlan, Indianapolis, IN) were used for the experiments at 13 to 14 wk of age. Animals were housed in a controlled environment and had free access to standard rodent diet (Harlan Teklad, Madison, WI) and tap water until the day before the experiments. All procedures were approved by the institutional animal care and use committee.
Blood pressure measurements in awake animals.
Male lean and obese ZR were anesthetized with isoflurane and fitted with chronic indwelling femoral arterial catheters (PE-50 tubing filled with heparinized isotonic saline solution), as previously detailed (28). Catheters were tunneled subcutaneously to the nape of the neck and exited via an 18-gauge needle wound and sealed with a steel pin until use. After a 3-day surgical recovery period, inline blood pressure was measured daily using a pressure transducer (TSD 104A, Biopac Systems, Santa Barbara, CA) coupled to a polygraph system (Biopac Systems) and a personal computer. Animals were then treated with the HO inhibitor ZnDPBG (25 μmol·kg−1·day−1 ip.) or matched vehicle (2 ml·kg−1·day−1 50 mmol/l Na2CO3 ip.) for 3 days. This particular dose of ZnDPBG was chosen to decrease but not maximally inhibit HO-derived CO formation (52).
Determination of whole animal CO excretion.
Another group of age-matched awake lean and obese ZR that did not receive the surgery were treated with the HO inhibitor ZnDPBG (25 μmol·kg−1·day−1 ip) or matched vehicle (2 ml·kg−1·day−1; 50 mmol/l Na2CO3 ip) for 3 days for daily measurement of CO excretion. Animals were placed in an acrylic airtight chamber with the outflow leading to a heated mercuric oxide bed coupled with a gas chromatograph (Peak, Mountain View, CA) for the determination of CO concentration, as has been detailed elsewhere (54, 55). The chamber was continuously purged with purified air and the outflow sampled for CO concentration at 2-min intervals. After 10 min of equilibration, the average of four measurements were used to calculate the CO excretion rate for the whole animal.
Metabolic parameters measurements and tissue extractions.
On the day of the experiment, rats were weighed and anesthetized with thiobutabarbital (Inactin, 120 mg/kg ip), and a carotid arterial catheter was implanted for blood sample collections. Blood samples were drawn for immediate determination of nonfasting blood glucose (Accu-Chek Compact, Roche Diagnostics, Indianapolis, IN), glycated hemoglobin (HbA1c, DCA 2000+ analyzer, Bayer, Pittsburgh, PA), and carboxyhemoglobin (HbCO) levels (OSM3 analyzer, Radiometer America, Westlake, OH), and lipid profile (Cardio Chek PA analyzer, QAS, Orlando, FL). Additional blood was collected in test tubes containing EDTA, and plasma was collected by centrifugation. Plasma samples were aliquoted and stored at −20°C until analyzed. Plasma insulin (rat insulin ELISA kit; Cayman Chemical, Ann Arbor, MI) and oxidized LDL (competitive oxidized LDL ELISA kit, Mercodia, Winston Salem, NC) were later determined. Animals were then heparinized (1000 U/kg iv), and the heart, right kidney, thoracic aorta, and the gracilis anticus muscles were removed and placed into ice-cold modified Krebs buffer. Right kidney and heart wet weights were then determined.
Thoracic aorta segments were removed as described above, snap-frozen in liquid N2, and stored at −70°C until analyzed. Aorta samples were pulverized in liquid N2 using a mortar and a pestle. HO-1 protein concentration was determined using the Stressgen rat HO-1 ELISA kit (Stressgen, San Diego, CA), and protein concentration was determined using a micro BCA protein assay kit (Pierce Biotechnology, Rockford, IL). HO-1 content was expressed as nangograms of HO-1 protein per milligram of total protein.
Isolated microvessel experiments.
Segments of first-order gracilis muscle arterioles were isolated by microdissection (45) and cannulated at both ends with glass micropipettes in a vessel chamber (Living Systems Instrumentation, Burlington, VT). The vessel chamber was continuously superfused with gased buffer (14% O2-5% CO2-balanced with N2; 37°C) via a nonrecirculating system. For internal diameter measurements, the vessel chamber was mounted on a stage of an inverted microscope (Nikon TS 100-F) fitted with a CCD video camera. The camera was connected to a personal computer equipped with video dimensioning software (ImagePro Express, Media Cybernetics). With this setup, a magnified image of the arteriolar segment was performed on the computer monitor, and the internal diameter was measured by manually adjusting guides superimposed by the video-dimensioning software. The software collected images at 1 frame/s that were stored as digital video files for documentation.
For experiments with no luminal flow, the proximal micropipette was connected to a pressure servo controller (Living Systems Instrumentation), and the distal micropipette was connected to a closed stopcock to achieve and maintain 80 mmHg constant luminal pressure with no flow. After a 60-min stabilization period, the HO inhibitor, 15 μmol/l CrMP, or vehicle was included in the superfusion buffer 20 min before the experiment. This treatment regime was continued throughout the experiment. After the pretreatment period, increasing concentrations of the endothelium-dependent vasodilator, ACh (1 nmol/l-3 μmol/l), were added to the superfusion buffer. Each concentration was tested for 5 min, internal diameter was recorded every minute, and the average of the last two measurements was used to determine the response.
To study flow-induced dilation, both the proximal and distal micropipettes were connected to pressure servo controllers and to an inline micro flowmeter (Living Systems Instrumentation). The HO inhibitor 15 μmol/l CrMP alone, the HO inhibitor, the HO product CO (mean system concentration 100 μmol/l), or vehicle was included in the luminal perfusion buffer. During a 60-min stabilization period, both proximal and distal pressures were adjusted to 80 mmHg with no luminal flow. During the experiments, proximal and distal pressures were adjusted equally in opposite directions to maintain the midline pressure at 80 mmHg and to establish graded levels of luminal flow (0–50 μl/min in 5μl/min increments). Each flow was tested for 5 min, internal diameter was recorded every minute, and the average of the last two measurements was used to determine the response. CO concentration was determined at the inflow and outflow ports of the microvessel micropipettes, and mean concentration was calculated for the system. In these flow experiments, drugs were only delivered via the luminal perfusion. CO concentration decreased from 200 μmol/l to 0 μmol/l while passing through the microvessel, most likely because of diffusion through the thin arteriolar wall.
All data are expressed as means ± SE. Blood pressure and CO excretion measurements, and vascular responses were analyzed by ANOVA using a computer statistical package (Sigma Stat 3.0). When significant differences were observed, orthogonal contrasts were performed as a post hoc analysis (44). All other data were analyzed by t-tests or when equal variance or normality tests failed by Mann-Whitney rank sum tests. A value of P < 0.05 was considered statistically significant.
Blood pressure and CO excretion measurements.
Awake obese ZR had higher blood pressures (P < 0.05), respiratory CO excretion rates (P < 0.05), and HbCO levels compared with lean ZR. In obese ZR, administration of the HO inhibitor ZnDPBG (25 μmol·kg−1·day−1 ip) lowered mean arterial pressure without affecting heart rate (Fig. 1A). In obese ZR, the HO inhibitor ZnDPBG (25 μmol·kg−1·day−1 ip) also lowered CO excretion rates (Fig. 1A). In age-matched lean ZR, the HO inhibitor did not affect blood pressure (Fig. 1B), or CO excretion rates (Fig. 1B).
Body and organ weight measurements.
Table 1 summarizes the body, kidney and heart weights of lean and obese ZR after 3 days of treatment with the HO inhibitor, ZnDPBG (25 μmol·kg−1·day−1 ip) or matched vehicle. Obese ZR had higher body weights, as well as kidney and heart weights compared with lean ZR. ZnDPBG treatment did not affect body or organ weights in either group.
Metabolic parameter measurements.
Table 2 summarizes metabolic parameters in lean and obese ZR after 3 days of treatment with the HO inhibitor, ZnDPBG (25μmol·kg−1·day−1 ip) or matched vehicle. Nonfasting blood glucose, HbA1c, as well as insulin levels were higher in obese ZR compared with lean ZR, and ZnDBPG treatment did not affect these levels in either group. Total cholesterol and LDL cholesterol, but not HDL cholesterol levels were also higher in obese ZR. ZnDPBG treatment did not affect total or HDL cholesterol levels in either group. However, ZnDPBG treatment lowered LDL cholesterol levels in obese, but not in lean ZR. Oxidized LDL levels were also higher in obese ZR, and ZnDPBG did not affect these levels in either group. Triglyceride levels were also elevated in obese ZR and exceeded the upper measurement limit of the Cardio Check PA meter in every case. ZnDPBG treatment did not significantly affect triglyceride levels in either group.
Figure 2 shows HO-1 protein levels in thoracic aortas isolated from lean and obese ZR after 3 days of treatment with the HO inhibitor, ZnDPBG (25 μmol·kg−1·day−1 ip) or matched vehicle. Vascular HO-1 protein levels were not significantly different between lean and obese ZR. ZnDPBG treatment promoted an approximately twofold increase in HO-1 protein levels in both groups.
Isolated microvessel experiments.
First-order gracilis muscle arterioles isolated from obese ZR had larger passive internal diameters (lean: 198 ± 4 μm, n = 14 vs. obese: 210 ± 3 μm, n = 21). During the 60-min equilibration period, diameters of both lean and obese ZR arterioles decreased spontaneously. Although the active internal diameters were larger in obese ZR arterioles (lean: 142 ± 7 μm vs. obese: 159 ± 4 μm), there was no difference in myogenic tone between the two strains (lean: Δmax −62 ± 7 μm vs. obese: Δmax −56 ± 4 μm). Twenty-minute pretreatment with the HO inhibitor, 15 μmol/l CrMP, promoted similar vasoconstriction in both groups (lean: Δmax −37 ± 18 μm vs. obese: Δmax −40 ± 4 μm).
An endothelium-dependent vasodilator, ACh (1nmol/l–3 μmol/l), promoted concentration-dependent vasodilation in arterioles isolated from lean and obese ZR (Fig. 3A). However, ACh-induced vasodilation was greatly attenuated in obese ZR arterioles compared with the lean group (lean: Δmax 62 ± 7 μm; n = 5 vs. obese: Δmax 32 ± 4 μm; n = 8; P < 0.05). Acute in vitro pretreatment with an inhibitor of HO, 15 μmol/l CrMP, enhanced maximal responses in obese ZR arterioles and abolished the difference between lean and obese groups (lean: Δmax 58 ± 13 μm; n = 5 vs. obese: Δmax 59 ± 8 μm; n = 7) (Fig. 3B).
Increases in luminal flow (0–50 μl/min) promoted vasodilation in arterioles isolated from lean ZR but not in arterioles from obese animals (lean: Δmax 22 ± 2 μm; n = 4 vs. obese: Δmax −3 ± 2 μm; n = 6; P < 0.05) (Fig. 4A). Acute in vitro pretreatment with an inhibitor of HO, 15 μmol/l CrMP, restored flow-induced responses in obese ZR arterioles and abolished the difference between lean and obese groups (lean: Δmax 20 ± 2 μm; n = 5 vs. obese: Δmax 21 ± 2 μm; n = 6) (Fig. 4B). However, acute in vitro simultaneous pretreatment with the HO inhibitor, 15 μmol/l CrMP, and the HO product, 100 μmol/l CO, prevented the restoration of flow-induced dilation in obese ZR arterioles (Δmax −1 ± 3 μm; n = 4) (Fig. 4B).
This study is the first to document that endogenous CO production is increased in obese ZR with metabolic syndrome, and administration of a HO inhibitor lowers respiratory CO excretion, as well as blood pressure in awake, obese ZR. Endothelium-dependent vasodilator responses are decreased in skeletal muscle arterioles isolated from obese ZR. Furthermore, acute in vitro treatment with a HO inhibitor restores endothelial function of resistance vessels to lean ZR levels, but exogenous CO prevents these effects. These findings suggest that heme-derived CO formation is increased in obese ZR with metabolic syndrome and contributes to hypertension and endothelial dysfunction of resistance vessels. Previous studies suggested that formation of HO-derived CO, an established inhibitor of endothelial nitric oxide production, is increased during salt-sensitive forms of hypertension and contributes to arteriolar endothelial dysfunction and hypertension (21, 22, 48). The current study extends those findings into an experimental model of metabolic syndrome.
Because of a nonfunctional leptin receptor gene (19), the obese ZR is a well-established genetic model of obesity (58) and metabolic syndrome (29, 13). Consistent with this, we found that obese ZR exhibit all elements of metabolic syndrome, such as increased body weight, elevated blood glucose, insulin, triglycerides, total, as well as LDL cholesterol levels, and hypertension. Metabolic syndrome represents a major risk factor for the development of type 2 diabetes. A previous study reported that patients with type 2 diabetes show increased respiratory CO levels (39). Although the blood glucose levels in these obese ZR are not as high as in other experimental diabetes models, we did find that respiratory CO excretion, as well as HbCO levels, is increased in obese ZR, indicative of increased endogenous CO formation.
The primary endogenous source of CO formation is the enzymatic breakdown of heme by HO (47). Our previous studies suggested that increased HO-derived CO formation contributes to salt-induced hypertension in Dahl salt-sensitive rats (21, 48). This prompted us to examine the contribution of heme-derived CO formation to hypertension in obese ZR. We chose the HO inhibitor ZnDPBG because of its quick tissue distribution (52) and low toxicity. In Sprague-Dawley rats, systemic administration of larger doses (45 μmol·kg−1·12 hr−1 ip) of ZnDPBG increased blood pressure (28) because of decreased central nervous system CO formation and the subsequent increase in sympathetic outflow (25). In the current study, we chose a lower dose of ZnDPBG (25 μmol·kg−1·24 hr−1 ip), which decreases respiratory CO excretion (52) but minimizes central nervous system effects (52). We found that administration of a HO inhibitor, ZnDPBG, reduced respiratory CO excretion and lowered blood pressure in awake, obese ZR. These data suggest that increased HO-derived CO production might contribute to hypertension in obese ZR.
A potential mechanism by which the HO inhibitor could lower the blood pressure in obese ZR is through the improvement of metabolic parameters. Although ZnDPBG treatment did lower LDL cholesterol levels in obese ZR, body weight, as well as the rest of the other metabolic parameters (blood glucose, HbA1c, plasma insulin, total cholesterol, HDL cholesterol, and triglycerides levels) were not significantly affected by the HO inhibitor. These data argue against major metabolic mechanisms for the blood pressure-lowering effect of HO inhibition.
Our previous studies in Dahl salt-sensitive rats suggested that increased heme-derived CO production contributes to salt-induced hypertension (48) by promoting endothelial dysfunction in resistance vessels (21, 48). Decreased endothelium-derived nitric oxide-mediated vasodilation in skeletal muscle arterioles has been suggested to contribute to hypertension in obese ZR (12–14). Thus we decided to examine whether HO-derived CO contributes to arteriolar endothelial dysfunction in obese ZR. We similarly found that responses to an endothelium-dependent vasodilator, ACh, were attenuated in first-order gracilis muscle arterioles isolated from obese ZR. Although ACh is widely used to assess nitric oxide-mediated endothelial function, previous studies suggested that pathological states can impair muscarinic receptor signaling proximal to nitric oxide synthase (46). An alternative method to circumvent this potential feature is to generate nitric oxide by increasing shear forces along the vascular endothelium. We found that skeletal muscle arterioles from obese ZR failed to dilate in response to flow. Because both ACh and flow-induced (26) dilation can be abolished by nitric oxide synthase inhibition, these data suggest that obese ZR show impaired endothelium-dependent nitric oxide-mediated vasodilation.
To test whether HO contributes to endothelial dysfunction in obese ZR, we used another inhibitor, CrMP. This metalloporphyrin is a selective (3) and photostable (51) inhibitor of HO activity, which was reported to decrease CO formation in rat first-order gracilis muscle arterioles (57). We found that acute in vitro application of this HO inhibitor restored ACh and flow-induced vasodilator responses in obese ZR and abolished the differences between lean and obese ZR arterioles. These data suggest that a vascular HO product contributes to endothelial dysfunction in obese ZR.
Recent studies indicate the physiologically relevant concentrations of CO (0.5–50 μmol/l) (54, 55) can inhibit nitric oxide synthase (37, 56), promote endothelial nitric oxide synthase-dependent vasoconstriction (23, 42), and impair endothelium-dependent vasodilation (49). We have previously reported that heme-derived CO impairs flow-induced dilation in normotensive rat arterioles (26). We have now found that CrMP, a HO inhibitor, restores flow-induced dilation in arterioles from obese ZR. To test whether HO-derived CO mediates endothelial dysfunction in obese ZR arterioles, we used simultaneous treatment with the HO inhibitor and exogenous CO. CO prevented the restoration of endothelial function by the HO inhibitor, suggesting that HO-dependent endothelial dysfunction in obese ZR arterioles is most likely mediated by CO.
The two major isoforms of HO are the inducible HO-1 and the constitutive HO-2 (35). Pathological conditions (35), such as oxidative stress (2), diabetes (17), and salt-sensitive hypertension (21, 22, 48), induce HO-1 expression, which can lead to enhanced CO formation (38). However, HO-derived CO formation is normally limited by the availability of heme substrate (27), and HO activity can increase in the absence of enhanced HO-1 expression (38, 54). In the current study, we did not find differences in thoracic aorta HO-1 protein levels between lean and obese ZR. However, the lack of an increase in total vascular HO-1 protein content does not detract from pathophysiological significance of the elevated in vivo production of CO.
The effects of increased HO-1 activity may not be exclusively adverse, as numerous studies have emphasized the cardiovascular protective effects of HO-1 induction (10). Certainly, HO-1-derived CO can exert antiproliferative effects in vascular smooth muscle cells, and it serves an important protective role against neointima formation after balloon injury or atherosclerosis (9). Other studies have demonstrated the direct vasodilator effects of CO on vascular smooth muscle (15). In fact, we were the first to demonstrate the major contributions of endogenous CO formation to the maintenance of basal resistance vascular diameter (32). It is important to clarify that this vasodilator component of CO is maximal in the absence of a functional nitric oxide system, such as after mechanical denudation of the endothelium (balloon injury, atherosclerosis, or experimental removal) or inhibition of nitric oxide synthesis as in our initial studies (32). Our current study does not address the potential benefits of increased HO activity in the obese ZR but rather clarifies that such benefits may be at least partially undermined by its tendency to promote endothelial dysfunction and hypertension.
We have avoided some common manipulations of the HO system. Some studies have used a commercially available heme preparation (heme-l-arginate) to induce HO-1 and increase HO activity (30). Although heme-l-arginate does increase heme-derived CO formation, it also contains an abundance of the nitric oxide synthase substrate, l-arginine (1:8 = heme-l-arginine ratio). Because heme is not covalently bound to the l-arginine and because l-arginine and CO compete for binding to nitric oxide synthase (37), the excess l-arginine confers protection against the inhibitory effects of CO on nitric oxide synthase (23). Another increasingly popular experimental approach is to induce HO-1 expression using various metalloporphyrins. For example, a recent paper has shown the protective effects of cobalt protoporphyrin (CoPP) treatment on endothelial function in a model of type 1 diabetes (50). Although CoPP—like many other metalloporphyrins (35)—does increase HO-1 protein levels, it is also a potent inhibitor of HO activity (36). In fact, our preliminary studies show that, despite increased HO-1 expression, systemic administration of CoPP lowers respiratory CO excretion and tissue CO content. Thus the protective effects of CoPP treatment in diabetes may be due to decreased endogenous CO production, just as in our current study. Our current study shows that although ZnDPBG treatment— just as treatment with CoPP and other metalloporphyrins HO inhibitors (35)—increases vascular HO-1 protein levels in obese ZR, respiratory CO excretion is, in fact, decreased in these animals. This also underscores the importance of using in vivo assessments of enzyme activity, instead of solely relying on HO-1 expression as a marker of endogenous CO production.
Previous studies suggested two other major mechanisms for decreased endothelium-derived nitric oxide-mediated vasodilation in obese ZR arterioles. Oxidative stress has been documented in obese ZR (13), and it has been suggested to promote endothelial dysfunction by directly chelating nitric oxide (13, 14). However, oxidative stress is also a potent inducer of HO-1 expression (2) and also promotes the formation of oxidized LDL (7), which can promote endothelial dysfunction (7), as well as induce HO-1 expression (1). In the current study, we found that plasma-oxidized LDL levels are elevated in obese ZR. Therefore, the possibility exists that oxidative stress promotes endothelial dysfunction not just by direct chelation of nitric oxide, but also by increasing vascular HO-derived CO formation. The other major mechanism for decreased endothelium-derived nitric oxide-mediated vasodilation in obese ZR arterioles is mediated by the betaII isoform of PKC (5, 6). Recently, PKC has been shown to increase HO activity (4, 34), as well as HO-1 levels (34). Therefore, the possibility exists that part of the effects of PKC on endothelial nitric oxide production are mediated by increased HO-derived CO production. Thus, with our current results, we do not propose an alternate mutually exclusive mechanism for endothelial dysfunction in obese ZR, but rather identify a common mediator downstream from the other previously described mechanisms. This might also explain why other treatments, such as reducing oxidative stress by SOD, only confer partial restoration of endothelial function (12–14).
In summary, we found that endogenous CO production is increased in obese ZR with metabolic syndrome and that administration of a HO inhibitor to “normalize” CO excretion lowers blood pressure in awake, obese ZR. In addition, we find that endothelium-dependent vasodilatory responses are decreased in skeletal muscle arterioles isolated from obese ZR. Furthermore, acute in vitro treatment with a HO inhibitor restores endothelium-dependent dilation to lean ZR levels, but exogenous CO prevents these effects. These findings suggest that heme-derived CO formation is increased in obese ZR with metabolic syndrome and contributes to hypertension and endothelial dysfunction in resistance vessels. Our results may help to identify novel therapeutic targets to improve endothelial function and treat hypertension in patients with metabolic syndrome.
This work was supported by National Institutes of Health grants: an Institutional Development Award (IdeA) Program P-20-RR017659 (F. K. Johnson) from the National Center for Research Resources, and National Heart, Lung, and Blood Institute Grants R01 HL76187 (R. A. Johnson), and R-01-HL59976 (W. Durante) and R-01-HL-74966 (W. Durante), and fellowship F32 HL-76001 (K. E. Jackson); and by the American Heart Association's Established Investigator Grant (W. Durante).
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2006 the American Physiological Society