Mapping of ATP, Glucose, Glycogen, and Lactate Concentrations Within the Arterial Wall
Objective— In large- and medium-sized arteries, the diffusion distances for oxygen and nutrients are long. This has been suggested to make these vessels prone to develop local energy metabolic deficiencies that could contribute to atherogenesis. To validate this hypothesis, we introduced a new method to measure energy metabolites within the arterial wall at high spatial resolution.
Methods and Results— Bioluminescence imaging was used to quantify local metabolite concentrations in cryosections of snap frozen (in vivo) and incubated pig carotid artery rings. Incubation at hypoxia resulted in increased lactate concentrations, whereas ATP, glucose, and glycogen concentrations were decreased, especially in the mid media, where concentrations of these metabolites were close to zero. In snap frozen arteries, glycogen concentrations were markedly higher in deep layers of the media than toward the lumen. ATP, glucose, and lactate were more homogenously distributed.
Conclusions— Bioluminescence imaging is a new and powerful tool to assess arterial wall energy metabolism at high spatial resolution. Our experiments demonstrate heterogeneous distributions of energy metabolites under hypoxic in vitro conditions. Furthermore, we show that glycogen concentrations are higher in deep medial layers in vivo. This might represent a local adaptation to a low-oxygen microenvironment.
The arterial wall is supplied with oxygen and nutrients by diffusion and convection from luminal blood and from vasa vasorum in the adventitia and outer media. Metabolic waste products are removed by diffusion in the opposite direction. Diffusion distances in large- and medium-sized arteries are often close to, or even exceed, the 100 to 200 μm that is frequently stated as the diffusion limit for oxygen.1–4 Based on these facts, it has been suggested that the arterial wall is predisposed to develop local energy metabolic disturbances and that this is a key factor both in the initiation and progression of atherosclerosis.1 In atherosclerosis-prone areas, both insufficient transport5,6 and increased consumption of metabolites7 could contribute to local disturbances. Later in the disease process, increased diffusion distances, attributable to intimal thickening as well as a high consumption of oxygen8 and glucose9,10 by foam cells, could additionally aggravate energy metabolic predicaments and contribute to the development of a necrotic core. Supportive of this hypothesis, hypoxic zones have been demonstrated in situ in the normal pig arterial wall11 as well as in vitro12–14 and in vivo15 in rabbit atherosclerotic plaques. However, ATP production may be maintained under hypoxic conditions via anaerobic breakdown of glucose and glycogen, but this could lead to a potentially harmful accumulation of lactate. Therefore, other aspects of energy metabolism besides the supply of oxygen need to be analyzed. To the best of our knowledge, bioluminescence imaging is the only method available that allows analysis of ATP, glucose, and lactate concentrations at the necessary spatial resolution.16 We have recently adapted this method to measure local ATP concentrations within the arterial wall.14 In the present study, our methodology is extended to include determinations of glucose, glycogen, and lactate. The use of bioluminescence imaging to measure glycogen concentrations is presented for the first time in this article.
The high spatial resolution of bioluminescence imaging allows assessment of the energy metabolic situation at defined histological locations in different parts of the arterial tree. It is in essence an in vitro method, but it may be used to reflect the in vivo situation when snap frozen tissue is analyzed. It thus constitutes a new and powerful tool that for the first time allows concentrations of arterial wall energy metabolites to be measured at a microscopic level.
Snap Freezing and Incubation of Tissue
Experiment Set Up
A series of methodological experiments was conducted to optimize measurement conditions. The main aspects of these experiments are described in the online supplement, available at http://atvb. ahajournals.org.
Snap Frozen Tissue: In Vivo
Snap frozen pig carotid arteries were used to evaluate the in vivo situation. The same tissue was also used to analyze the methodological variation of bioluminescence imaging.
In Vitro Experiments
Two different sets of in vitro experiments were performed where pig carotid arteries were exposed to hypoxia and glucose deprivation at varying degree. In the first set of experiments, incubation was performed for 3 hours. The purpose of these experiments was to study the relation between energy metabolic status (ATP), the availability of energy substrates (glucose and glycogen), and the production of waste product (lactate) under varying oxygen and glucose concentrations. In the second set of experiments, substrate-free medium and varying incubation times (0 to 6 hours) were used. The purpose was to study the glycogen breakdown rate under various oxygen concentrations and relate this to the energy metabolic status (ATP) in the tissue.
The study protocol was approved by an Animal Ethics Committee and conducted in accordance with NIH guidelines for use of experimental animals. Carotid arteries were obtained from domestic pigs (51 to 74 kg). The animals were premedicated with 4:1 ketamine (Ketalar 100 mg/mL, Parke-Davis) and midazolam (Dormicum 5 mg/mL, Roche) 0.5 mL/kg, and anesthesia was induced with 0.07 to 0.14 mL of thiopenthal sodium IV (pentothal natrium 60 mg/mL, Abbot). After intubation, anesthesia was maintained with isoflurane (Isofluran Pharmacia, Pharmacia & Upjohn) in N2O/O2.
The right carotid artery was surgically exposed, and 1 1-cm-long segment was rapidly snap frozen in liquid nitrogen. Freezing was done within 15 seconds from induction of cardiac arrest (KCl IV) and within 2 to 3 seconds from removal. An additional a 6-cm-long segment to be used for in vitro incubations was put into oxygenated (75% O2, 5% CO2 in N2) PBS (without calcium and magnesium) with 5.6 mmol/L glucose.
PBS (without calcium and magnesium) with 25 mmol/L HEPES and 10 mmol/L bicarbonate with 0 or 5.6 mmol/L glucose was used throughout.
Dissection and Incubation
In the 6-cm segments used for in vitro studies, the adventitial tissue was carefully removed by dissection in circulating oxygenated (75% O2, 5% CO2 in N2) PBS at 37°C, and the vessel was divided into 5-mm rings and distributed into incubation vials.
In the first set of experiments (3 independent experiments), arterial segments were subjected to different degrees of energy metabolic restriction by incubation for 3 hours at different oxygen (75% or 3%) and glucose (5.6 or 0 mmol/L) concentrations at 37°C. In the second set of experiments (2 independent experiments), arterial segments were allowed to exhaust their glycogen stores by incubation in glucose-free PBS at 75% or 3% oxygen for different time periods (0, 0.5, 1.5, 3, and 6 hours) at 37°C.
Immediately after surgical removal or incubation, the aortic rings were snap frozen in liquid nitrogen, mounted in OCT (Tissue-Tek) and stored at −70°C until further use. From each ring, 24 consecutive cryosections (10 μm) (Leica cryostat at −25°C) were mounted on poly-l-lysine slides (Sigma). The sections were immediately fixed on a heating plate (95°C) for 10 minutes and then stored at −20°C until analysis. Additional 10-μm sections were taken from 10 artery rings (5 snap frozen, 5 from in vitro experiments). These sections were air dried overnight and subsequently stained with Mayer’s hematoxylin and used to determine cellularity profiles.
The cell nuclei in 10-μm sections from 10 artery rings were stained with Mayer’s hematoxylin. From each ring, digital images of 4 lumen to adventitia profiles (at 3, 6, 9, and 12 o’clock) were obtained. Each profile was a composite of 3 to 4 visual fields with a ×40 objective (Zeiss Axioplan 2 microscope and an Axiocam cooled CCD color camera, Carl Zeiss). The profiles were dived into 100-μm-thick zones based on the distance from the lumen and the areas, and number of nuclei in each of these zones was calculated (KS 400 software, Carl Zeiss).
Bioluminescence imaging was used to visualize and quantify the distribution of metabolites within cryosections. A range of methodological experiments was performed to optimize spatial resolution and the quantitative calibration as well as to define methodological variation of this method. A description of these experiments as well as detailed methodological instructions are given in the online supplement. A brief methodological description is given below.
An enzyme solution containing luciferase is applied (in complete darkness) to a cryosection on a microscope stage. This leads to emission of photons in proportion to the concentration of the studied metabolite (ATP, glucose, glycogen, or lactate). The photons are registered by a photon-counting camera (C2400-47, Hamamatsu Photonics) mounted on the microscope (Axiovert 135 M, Carl Zeiss). The light intensity (gray value) in different parts of the resulting digital bioluminescence image reflects the local metabolite concentrations. A darkfield image of the same section is also obtained to outline histological structures in the corresponding bioluminescence image. From every artery ring, 5 or 6 consecutive sections were analyzed for each of the 4 metabolites. Each section was used for the analysis of 1 metabolite, and all measurements were performed at room temperature (23±1°C).
To calibrate the bioluminescence signal, standards were made by dissolving different concentrations of ATP, glucose, glycogen, and lactate in physiological saline with 8% low-molecular-weight gelatin. The solution was frozen, and 10-μm cryosections were made and treated in the exact same way as the tissue sections.
To obtain values for the mean metabolite concentration within the whole media (ie, global metabolite concentration), the mean bioluminescence intensity (gray value) in 5 to 6 consecutive sections was calculated. To evaluate differences in metabolite concentrations at different distances from the lumen, concentration profiles were determined. Each profile is an average of the bioluminescence intensity along 12 virtual sections at 3, 6, 9, and 12 o’clock in 3 consecutive sections. Areas where small air bubbles had been trapped in the enzyme solution were excluded from measurement. KS 400 software (Carl Zeiss) was used throughout.
Determination of Methodological Variation
Snap frozen pig carotid artery segments from 5 animals were used to define methodological variation. From every artery segment, 24 sections were analyzed, ie, 6 each for the metabolites ATP, glucose, glycogen, and lactate.
Variability Between Consecutive Sections
Variation in bioluminescence intensity between measurements on 6 consecutive sections was analyzed in each of the 5 artery segments (ie, 5×6=30 sections per metabolite).
Variability Within Single Sections
All metabolites showed a homogenous distribution pattern along the circumference of the artery. Therefore, variation within sections could be assessed by determining mean metabolite concentrations in the 4 analog segments of each section (ie, 5×6×4=120 quadrants per metabolite).
Coefficients of variation and global metabolite concentrations are expressed as mean±SD. ANOVA with Bonferroni correction was used to compare groups in Figure 3, and P<0.01 was considered significant.
Consecutive sections from snap frozen segments of pig carotid artery (n=5) were used to assess methodological variability. The coefficient of variation between repeated measurements on consecutive sections was ≈10% for ATP (10±2%) and lactate (10±3%), whereas it was approximately twice as high for glucose (18±3%) and glycogen (20±5%). The variation within analog segments of a single section was lower, and the coefficient of variation ranged from 7±4% for ATP and lactate to 12±6% for glycogen (glucose, 9±5%).
The ATP-, glucose-, glycogen-, and lactate-dependent bioluminescence intensity in 10-μm frozen sections with known metabolite concentrations is shown in Figure 1. The standard curves were subsequently used to transform bioluminescence intensity (relative units) in tissue sections into actual metabolite concentrations (micromolar per gram of wet weight). In the concentration range found within the tissue sections, a linear approximation could be used to calculate metabolite concentrations.
Cellularity profiles were determined to exclude the possibility that different metabolite concentrations were attributable to differences in cellularity between different medial layers. Cellularity was ≈20% lower toward the adventitia (2100±200 cells/mm2) than toward the lumen (2500±240 cells/mm2) (Figure 2). There was no difference in the cellularity profiles between snap frozen arteries and arteries that had been used for in vitro experiments.
In Vivo: Snap Frozen Arteries
To evaluate the in vivo situation, snap frozen segments of pig carotid arteries (n=5) were analyzed.
Global Metabolite Concentrations
The global metabolite concentrations in snap frozen pig carotid arteries (n=5) are given in Figure 3 (black bars).
Concentration profiles were assessed to detect heterogeneities in the concentration of metabolites at different depths of the tissue (Figure 4). The concentration profiles indicated homogenous ATP and lactate concentrations throughout the media. Glucose concentrations, however, were higher toward the lumen, declined to reach a nadir in the mid media (≈400 to 500 μm from the lumen), and increased again toward the adventitia. Glycogen distribution showed an opposite pattern, with higher concentrations in the mid media than toward the lumen and adventitia.
Effect of Incubation at Different Oxygen and Glucose Concentrations
Pieces of normal pig carotid artery were incubated for 3 hours at 37°C in medium with different oxygen (75% or 3% O2) and glucose (0 or 5.6 mmol/L) concentrations.
To emphasize changes, ATP, glucose, glycogen, and lactate concentrations are given in percents, with 100% corresponding to concentrations after incubation at 75% O2 and 5.6 mmol/L glucose. The absolute concentrations, in micromolar per gram of wet weight, are given in Figure 3. When oxygen supply was compromised (3% O2) at 5.6 mmol/L glucose in the medium, ATP concentrations were reduced to 49±7%. Glucose concentrations were reduced to 63±27%, and glycogen concentrations to 33±14%. Lactate concentrations, however, were increased to 145±32%. Thus, ATP concentrations were reduced by approximately half, despite the indication of an increase in anaerobic breakdown of glucose and glycogen to lactate under hypoxic conditions. When glucose was removed from the medium but oxygen supply was abundant, ATP concentrations were slightly reduced to 88±16%, and glycogen and lactate concentrations were reduced to 49±14% and 59±22%, respectively. As expected, glucose concentrations were very low, reduced to 5.1±4.7%. The ATP production in the absence of extracellular energy substrates can be maintained by utilization of intracellular energy stores, eg, glycogen or fat. Because glycogen concentrations were approximately 50% lower than when glucose was present during incubation, it is likely that the ATP production in this set up, at least in part, derived from breakdown of glycogen. When both oxygen and glucose supply were compromised, there was a pronounced reduction of ATP to 9.4±9.0%, glucose to 3.6±3.2%, and glycogen to 0.5±3.1%. Lactate was reduced to 53±18%. Thus, glycogen breakdown was more pronounced under hypoxic conditions, and depletion of available stores coincided with failed ATP production.
Determination of concentration profiles made it possible to detect heterogeneities in the concentration of metabolites at different depths of the tissue. At high oxygen and 5.6 mmol/L glucose, there was a glucose gradient with an approximately 20% lower glucose concentration in the mid media (≈400 μm from the lumen). Glycogen concentrations, on the other hand, were lower in the layers adjacent to the lumen than toward the adventitia. ATP and lactate were more homogenously distributed (Figure 5). Reduction of the supply of oxygen with glucose concentrations maintained had a pronounced effect on the concentration profiles. The concentrations of ATP, glucose, and glycogen were zero in the mid media, approximately 300 to 400 μm from the lumen, and higher toward lumen and adventitia. Thus, local substrate depletion in the mid media colocalized with failed ATP production. However, lactate concentrations were increased throughout the media and were not heterogeneously distributed (Figure 5). With abundant oxygen but without glucose, the ATP and glycogen concentrations were slightly higher toward the adventitia than in the luminal layers. Lactate and glucose concentrations were low, and these metabolites were more evenly distributed within the tissue (not shown). Reduction of the supply of oxygen (3%) as well as glucose resulted in ATP depletion in medial layers deeper than >100 to 200 μm from the lumen. However, in a 100- to 200-μm-wide zone adjacent to the lumen, ATP concentrations were higher, up to 0.10 μmol/g wet weight. Glucose, glycogen, and lactate concentrations were low and more homogenously distributed (not shown).
Effect of Incubation Without Energy Substrates for Different Time Periods
To exhaust the available intracellular glycogen stores, pieces of pig carotid artery were incubated in well-oxygenated (75% O2) or hypoxic (3% O2) glucose-free PBS for 0.5 to 6 hours. Zero hours (t=0) represents an artery segment frozen after dissection, immediately before incubation.
In incubations with high oxygen (Figure 6, solid lines), ATP concentrations were only slightly reduced after 3 hours. After 6 hours, however, ATP concentrations were close to zero. Glycogen concentrations showed a gradual decline and were very low after 6 hours. Glucose and lactate concentrations were reduced at all time points. In comparison with high-oxygen incubations, hypoxia (Figure 6, dashed lines) resulted in very low concentrations of ATP as well as glycogen already after 1.5 hours. Lactate concentrations, on the other hand, were higher after 0.5 hours but at a similar level as in well-oxygenated medium after prolonged incubation. Glucose concentrations were reduced at all time points and thus were not influenced by oxygen concentrations. The data indicate an anaerobic breakdown of glycogen under hypoxic conditions, with a more rapid depletion of available stores and subsequent failure of ATP production.
In a recent publication, we introduced a new technology to determine ATP concentrations within the arterial wall at high spatial resolution.14 In this study, our methodology is extended to include determinations of local glucose, glycogen, and lactate concentrations as well. Both snap frozen (in vivo) and in vitro incubated pig carotid arteries were analyzed. The results presented here lend themselves to 2 principally different discussions. First, the new methodology can detect local heterogeneities in the concentration of energy metabolites within the arterial wall. Second, the in vivo and in vitro data from our experiments add new insight into the heterogeneity of energy metabolism within the arterial wall.
The methodology is based on the registration of metabolite-specific bioluminescence in frozen tissue sections. In this study, the technique was used to determine global metabolite concentrations (average concentration in the whole media) and concentration profiles within the arterial wall. A range of established techniques can be used to measure global concentrations (ie, data as in Figures 3 and 6⇑) of energy metabolites. However, these methods do not provide any information about the regional distribution. The analyses performed in this study illustrate the importance of these regional assessments, because the metabolite distributions might well be heterogeneous (Figures 4 and 5⇑). The variability in bioluminescence intensity between analogous segments of single sections is low (coefficient of variation, 7% to 12%), which makes registration of false heterogeneities unlikely. If consecutive sections are used for the analysis of ATP, glucose, glycogen, and lactate, these key aspects of energy metabolism can be studied within the same microscopic location (Figures 4 and 5⇑). An additional strength is that additional sections may be used for histological staining, immunohistochemistry, and in situ hybridization.
Inherent to the presented method is diffusion of substrate from the frozen sections during bioluminescence registration.14,17 Optimization of measurement conditions (described in the online supplement) has reduced this artifact, but true tissue gradients will be slightly underestimated. This is most clearly illustrated at tissue borders, where an infinitely steep gradient would be expected but a less steep gradient is registered (profiles in Figures 4 and 5⇑).
To get a true reflection of the in vivo situation with the presented methodology, the time from cessation of blood flow until the tissue is frozen should be minimized. During the delay between removal and freezing, metabolite concentrations could change, eg, ATP would be expected to decline.18 Furthermore, diffusion of small solutes before freezing could lead to smoothening, and thus underestimation, of true tissue gradients. In our study, the time from cessation of blood flow to snap freezing was kept at an absolute minimum (between 2 to 3 seconds), and we believe that our results reflect the in vivo situation. However, it is impossible to rule out that in vivo metabolite concentrations could be subject to some change even when freezing is almost instantaneous.
Aerobic Lactate Production
The lactate determinations performed in our study indicated a substantial lactate production even after incubation in an excess of oxygen, where more efficient oxidative metabolism would have been expected to occur (Figures 3 and 5⇑). Glucose, and not glycogen, seemed to be the main substrate for this aerobic lactate production, because without glucose, lactate concentrations were lower despite remaining glycogen stores (Figure 3). Our findings support previous in vitro studies claiming that aerobic glycolysis is a characteristic feature of vascular smooth muscle cells19–21 and that glucose, and not glycogen, is the substrate used in this process.19 As an explanation of this phenomenon, it has been suggested that different metabolic pathways are compartmentalized within the vascular smooth muscle cells and that glycolysis is specifically coupled to sodium and potassium transport processes whereas oxidative metabolism is coupled to contractile energy requirements.19 This compartmentation, which is not absolute,22,23 has been suggested to lead to a more efficient coupling between energy supply and transduction. The finding that lactate concentrations were even higher in snap frozen tissue (Figures 3 and 4⇑) strongly suggests that the high lactate concentrations were not an in vitro artifact but rather a reflection of normal vascular smooth muscle metabolism.
Energy Metabolism Under Hypoxia
When the availability of oxygen in the surrounding medium was compromised, adequate ATP production could not be maintained. Glucose and glycogen concentrations were decreased and lactate concentrations increased, indicating a compensatory increase in anaerobic metabolism (Figures 3 and 5⇑), in agreement with other studies.21,24 This seemed to offer a certain degree of protection against hypoxia-induced energy metabolic failure, because ATP concentrations were much lower at low oxygen without glucose (Figure 3). In the mid media, ATP, glucose, and glycogen concentrations were close to zero (Figure 5). Identical in vitro conditions have previously been shown to induce severe hypoxia in this zone of the normal rabbit aorta.13,14 The pig carotid artery is approximately twice as thick as the rabbit aorta and would be expected to be close to anoxic in the mid media under the same experimental conditions. The shortage of oxygen would be expected to lead to an increase in anaerobic energy metabolism, consequently straining the available diffusion capacity for glucose as well as breakdown of the intracellular glycogen stores. Thus, the mid medial ATP depletion is probably secondary to the local deficiency of both oxygen and substrates in this zone.
A decrease in arterial wall oxygenation attributable to atherosclerosis,25,26 hypertension,27 or hypercholesterolemia28 leads to a compensatory increase in the number of adventitial vasa vasorum. This increases tissue perfusion and counteracts the development of arterial wall hypoxia in vivo.25 However, there seem to be a limit for this compensatory mechanism, because the core of rabbit atherosclerotic plaques is severely hypoxic in vivo when the intimal thickness exceeds 300 μm.15 This failure to adequately increase tissue perfusion might be explained by the more irregular growth pattern and poorly supported wall of the newly formed vasa vasorum.26,28
Glycogen as an Energy Substrate
The relative importance of different intracellular energy stores in vascular smooth muscle cells is controversial, and varying glycogen stores have been reported from different vessels.29–31 In this study, we show that ATP levels could be maintained without exogenous energy substrates, but only when sufficient glycogen stores were available (Figures 3 and 6⇑). Based on data from independent groups, it has previously been estimated that glycogen stores can maintain ATP production for 38 minutes in contracting pig carotid arteries.32 However, in our setup, ATP levels were maintained for 3 hours under well-oxygenated conditions. This might be explained by the fact that the vessel segments were unstimulated in our experiments and could thus be expected to have a substantially lower energy demand than contracting arteries. Glycogen depletion was more rapid and the initial lactate concentrations were higher under hypoxic conditions (Figure 6). This strongly suggests an anaerobic, and less-efficient, breakdown of glycogen under hypoxic conditions and, as a result, a more rapid depletion of the available glycogen stores.
Glycogen concentrations in snap frozen carotid arteries were higher in the deep layers of the media than close to the lumen (Figure 4). Consequently, glycogen concentrations were high in the layers where in vivo oxygen concentrations would be expected to be low.11 It has recently been shown that culture of rat tail arteries under hypoxic conditions leads to a very pronounced upregulation of the glycogen stores.29 A possible explanation of our finding could thus be that the hypoxic microenvironment in the mid media leads to a local upregulation of the glycogen stores in these layers. This adaptive mechanism could make these layers more tolerant to future episodes of hypoxia and nutrient shortage.
Local energy metabolic disturbances within the arterial wall have been proposed to be a key factor in atherogenesis. In this study, we introduce a novel methodology that allows concentrations of arterial wall energy metabolites to be measured at a microscopic level. With this new technique, pronounced heterogeneities in the concentrations of ATP, glucose, and glycogen under hypoxic in vitro conditions were demonstrated. Furthermore, we show for the first time that glycogen concentrations in pig carotid arteries are higher in deep medial layers than close to the lumen. The upregulation of glycogen stores in these layers might represent an adaptation of the smooth muscle cells to a low-oxygen microenvironment.
This study was supported by grants from the Swedish Heart Lung Foundation (project No. 51009, 61007, and 61510) and the Swedish Medical Research Council (No. 04531 to G.B. and No. 13488 to O.W.). The authors thank Karl Swärd, PhD and Anders Lindqvist, PhD, at Lund University, Sweden, for valuable advice and control tissues during the set up of the glycogen methodology.
- Received August 1, 2003.
- Accepted August 8, 2003.
Heuper WC. General reviews. Arch Pathol. 1944; 38: 162–181.
Wolinsky H, Glagov S. Nature of species differences in the medial distribution of aortic vasa vasorum in mammals. Circ Res. 1967; 20: 409–421.
Kosan RL, Burton AC. Oxygen consumption of arterial smooth muscle as a function of active tone and passive stretch. Circ Res. 1966; 18: 79–88.
Björnheden T, Bondjers G. Oxygen consumption in aortic tissue from rabbits with diet-induced atherosclerosis. Arteriosclerosis. 1987; 7: 238–247.
Rudd JH, Warburton EA, Fryer TD, Jones HA, Clark JC, Antoun N, Johnstrom P, Davenport AP, Kirkpatrick PJ, Arch BN, Pickard JD, Weissberg PL. Imaging atherosclerotic plaque inflammation with [18F]-fluorodeoxyglucose positron emission tomography. Circulation. 2002; 105: 2708–2711.
Barker SG, Talbert A, Cottam S, Baskerville PA, Martin JF. Arterial intimal hyperplasia after occlusion of the adventitial vasa vasorum in the pig. Arterioscler Thromb. 1993; 13: 70–77.
Björnheden T, Evaldsson M, Wiklund O. A method for the assessment of hypoxia in the arterial wall, with potential application in vivo. Arterioscler Thromb Vasc Biol. 1996; 16: 178–185.
Levin M, Björnheden T, Evaldsson M, Walenta S, Wiklund O. A bioluminescence method for the mapping of local ATP concentrations within the arterial wall, with potential to assess the in vivo situation. Arterioscler Thromb Vasc Biol. 1999; 19: 950–958.
Björnheden T, Levin M, Evaldsson M, Wiklund O. Evidence of hypoxic areas within the arterial wall in vivo. Arterioscler Thromb Vasc Biol. 1999; 19: 870–876.
Allard JR, Conhaim RL, Vlahakes GJ, O’Neill MJ, Hoffman JI. Rapid-freezing transmural cardiac biopsy drill. Am J Physiol. 1981; 240: H126–H132.
Lynch RM, Paul RJ. Compartmentation of glycolytic and glycogenolytic metabolism in vascular smooth muscle. Science. 1983; 222: 1344–1346.
Suzuki LA, Poot M, Gerrity RG, Bornfeldt KE. Diabetes accelerates smooth muscle accumulation in lesions of atherosclerosis: lack of direct growth-promoting effects of high glucose levels. Diabetes. 2001; 50: 851–860.
Ishida Y, Riesinger I, Wallimann T, Paul RJ. Compartmentation of ATP synthesis and utilization in smooth muscle: roles of aerobic glycolysis and creatine kinase. Mol Cell Biochem. 1994; 133–134: 39–50.
Kuwahara F, Kai H, Tokuda K, Shibata R, Kusaba K, Tahara N, Niiyama H, Nagata T, Imaizumi T. Hypoxia-inducible factor-1alpha/vascular endothelial growth factor pathway for adventitial vasa vasorum formation in hypertensive rat aorta. Hypertension. 2002; 39: 46–50.
Lindqvist A, Dreja K, Sward K, Hellstrand P. Effects of oxygen tension on energetics of cultured vascular smooth muscle. Am J Physiol Heart Circ Physiol. 2002; 283: H110–H117.
Odessey R, Chace KV. Utilization of endogenous lipid, glycogen, and protein by rabbit aorta. Am J Physiol. 1982; 243: H128–H132.
Allen TJ, Hardin CD. Influence of glycogen storage on vascular smooth muscle metabolism. Am J Physiol Heart Circ Physiol. 2000; 278: H1993–H2002.