Skip to main content
NCBI home page
The Journal of Physiology logoLink to The Journal of Physiology
. 2014 Feb 17;592(Pt 6):1225–1235. doi: 10.1113/jphysiol.2013.267302

Muscle contraction increases interstitial nitric oxide as predicted by a new model of local blood flow regulation

Aleksander S Golub 1, Bjorn K Song 1, Roland N Pittman 1,✉
PMCID: PMC3961083  PMID: 24445318

Abstract

The prevailing metabolic theory of local blood flow regulation suggests the dilatation of arterioles in response to tissue hypoxia via the emission of multiple metabolic vasodilators by parenchymal cells. We have proposed a mechanism of regulation, built from well-known components, which assumes that arterioles are normally dilated in metabolically active tissues, due to the emission of NO by the endothelium of microvessels. Regulation of local blood flow aims at preventing an excessive supply of oxygen (O2) and glucose to the tissue and thus provides an adequate supply, in contrast to the metabolic regulation theory which requires permanent hypoxia to generate the metabolic vasodilators. The mediator of the restrictive signal is superoxide anion (O2−) released by membrane NAD(P)H oxidases into the interstitial space, where it neutralizes NO at a diffusion-limited rate. This model predicts that the onset of muscle contraction will lead to the cessation of O2− production, which will cause an elevation of interstitial NO concentration and an increase in fluorescence of the NO probe DAF-FM after its conversion to DAF-T. The time course of DAF-T fluorescence in contracting muscle is predicted by also considering the washout from the muscle of the interstitially loaded NO indicator. Experiments using pulse fluorimetry confirmed an increase in the interstitial concentration of NO available for reaction with DAF-FM during bouts of muscle contraction. The sharp increase in interstitial [NO] is consistent with the hypothesis that the termination of the neutralizing superoxide flow into the interstitium is associated with the activation of mitochondria and a reduction of the interstitial oxygen tension. The advantage of the new model is its ability to explain the interaction of metabolic activity and local blood flow through the adequate delivery of glucose and oxygen.

Introduction

The prevailing metabolic theory of local blood flow regulation suggests that an increase in cell respiration leads to tissue hypoxia and then to the emission of vasodilator metabolites, which increase local blood flow (Roy & Brown, 1880; Rowell, 2004). However, this mechanism implies that the restoration of an adequate oxygen supply stops the production of the metabolic signal, which should reduce blood flow back to a state of tissue hypoxia. The drawback of the metabolic control theory is the maintenance of tissue hypoxia required for emission of metabolic vasodilators. A century long quest has not led to finding the specific metabolite, exported from parenchymal cells and serving as a vasodilator messenger. The well-established potent vasodilator NO is not suited for this role because it is generated by the endothelial cells of blood vessels and cannot report the metabolic demand of the parenchyma.

We have proposed a novel mechanism for local blood flow regulation based on the concept that the normal physiological state for the musculature is active contraction and the corresponding normal state for its vasculature is dilatation (Golub & Pittman, 2013). That state is supported through continuous production of NO by eNOS in endothelial cells (Figueroa et al. 2001). Active regulation of local blood flow comes into effect when the rate of oxygen and glucose supply reaches or exceeds the tissue demand, thus leading to an abundance of cytosolic reducing agents and extracellular oxygen that are the substrates for membrane NAD(P)H oxidase. This circumstance causes the emission of superoxide anion radical (O2−) into the interstitial space and subsequent neutralization of the interstitial NO with O2−, particularly in resting muscle. The onset of contractions and activation of mitochondria terminate the production of superoxide into the interstitial space by NAD(P)H oxidase localized in the sarcolemma. This is due to the import of reducing equivalents from the cytosol into the activated mitochondria and the fall of oxygen tension (Inline graphic) on the surface of muscle fibres. Reduction of superoxide emission into the extracellular space opens the interstitial space for the diffusive flux of NO to the smooth muscle cells in arterioles, causing them to dilate and increase local blood flow. Thus, the interstitial concentration of nitric oxide acts as an error signal whose magnitude is related to the mismatch between the delivery of oxygen and glucose to and their consumption by the skeletal muscle.

A feedback control circuit based on this principle is shown in Fig. 1. When oxygen and glucose content in the inflowing blood are sufficient, endothelial eNOS produces a constant flux of NO into the interstitial space, known as the basal production of NO. This loop provides a positive feedback for the smooth muscle cells (SMC) of the arteriolar wall. This NO flux serves as a reference signal and is constant if blood Inline graphic exceeds the dependent range on the oxygen dependence curve for eNOS (at least, twice larger than Km = 2.3 mmHg for this enzyme (Stuehr et al. 2004)). Transport of oxygen and glucose to skeletal muscle fibres is restricted by diffusion and transmembrane glucose transporters. This limitation is represented in Fig. 1 as a resistor for O2 and glucose flow. The local muscle cells are the consumers of glucose and oxygen and at the same time the sensors of their adequate supply. With sufficient or excessive delivery of glucose, the membrane NAD(P)H oxidase is supplied with its substrates NADH and NADPH from the cytosol to transfer electrons across the sarcolemma to the oxygen molecules and convert them into superoxide. Thus, the negative feedback signal reporting the achievement of a high level of supply is encoded as the rate of superoxide emission into the interstitium.

Figure 1.

Figure 1

Open in a new tab

The presentation is similar to the classical feedback control circuit consisting of a controller, the control system, the control process and the feedback loop. Microvascular endothelium, which is in contact with arterial blood of constant composition, produces a flow of NO (reference signal) into the perivascular interstitial space (IS). It is a positive feedback loop that provides a reference signal reporting the availability of O2 and glucose supply in blood. In the absence of superoxide in the interstitium, NO causes relaxation of the smooth muscle cells (SMCs) via the soluble guanylyl cyclase (sGC) pathway. SMCs in the arteriolar wall serve as a controller, while the inner diameter of the arteriole is the controlled system and local blood flow is the controlled process. The negative feedback loop includes the transport resistance to oxygen and glucose diffusion, together with the resistance of glucose membrane transporters. Muscle fibres (parenchymal cells) perform the function of the sensor, which produces superoxide flow with membrane NAD(P)H oxidase to the interstitium (negative feedback) when the cell's supply of glucose and oxygen is sufficient. Subtracting a feedback signal from the reference signal and forming an error signal as [NO] occurs in the interstitial space.

It is well known that superoxide anion is a specific antagonist of NO, and it reacts with NO at a rate limited by diffusion: from 6.7 × 109 m−1 s−1 (Beckman & Koppenol, 1996) to 1.9 × 1010 m−1 s−1 (Kissner et al. 1997). Rapid neutralization of NO by superoxide anion in the interstitium of resting muscle may be the underlying cause of the problem with determining the physiological concentration of NO in tissues (Hall & Garthwaite, 2009). Thus, the perivascular concentration of NO in the muscle is determined by the interaction of NO and O2− fluxes. The superoxide flux is dependent on a sufficient supply of O2 and glucose to parenchymal cells and is subtracted from the NO flux, determined by the high content of oxygen and glucose in blood. The result of this subtraction is the error signal, represented as the NO concentration at the SMCs of the arteriolar wall. These cells serve as the controller of vascular wall tone, changing the inner diameter and hydraulic resistance of arterioles to adjust the local blood flow (Fig. 1).

In this model the local regulator actively controls the upper level of the supply rates for both oxygen and glucose at the transition from exercise to resting conditions, so that the constriction of arterioles at rest prevents cells from being exposed to an excessive supply of oxygen and substrates. Thus, the goal of this type of regulation is a sufficient supply of oxygen and substrates to the cells. This is in opposition to the hypothetical metabolic regulation principle which postulates the production of vasodilator metabolites in response to a lack of oxygen supply, thus maintaining a state of tissue hypoxia. Moreover, the metabolic theory assumes that in hypoxic conditions parenchymal cells have sufficient resources and time for the accumulation of signalling metabolites to their acting concentrations. In contrast, the controller illustrated in Fig. 1 simply requires a termination of the inhibitory signal (O2−) to increase local blood flow. In addition, the presence of extracellular superoxide dismutase (ecSOD) significantly accelerates the removal of the residual superoxide, reducing the time for development of functional hyperaemia (Karlsson & Marklund, 1988).

Because of the relatively low rate of reaction of known fluorescent probes with NO, they cannot compete with superoxide in the reaction with NO (Namin et al. 2013). However, according to the proposed mechanism of NO–O2− regulation, the onset of muscle contractions rapidly reduces the superoxide production by skeletal muscle cells which elevates interstitial [NO] and makes it more available for reaction with a probe. In the case of DAF-FM loaded into the interstitial space, the rapidly increased availability of NO should cause a sharp increase in the concentration of the fluorescent triazole form DAF-T (Nagano, 2009). It is anticipated that superoxide withdrawal should be manifested as a rapid increase of the fluorescence intensity, synchronous with the onset of muscle contraction. Experimental observation of this predicted phenomenon would be a corroboration of the validity of the proposed mechanism for the local regulation of blood flow.

Methods

Experimental design

Termination of superoxide emission into the interstitial space in response to stimulated isometric muscle contractions is assumed to open the way for the reaction of NO with DAF-FM, to convert it into the highly fluorescent triazole form DAF-T (Nagano, 2009). In order to find optimal conditions for observation of the NO-related fluorescence and to predict the time course of the fluorescence intensity caused by a bout of muscle contraction, we also have to account for the effect of washout of both forms of the indicator (DAF-FM and DAF-T) from the interstitial space. The fluorescent probe DAF-FM cannot penetrate the cell membrane and is distributed only throughout the interstitial space of a muscle. The concentration of the low molecular weight dye in the interstitial space of a muscle decreases with time because of indicator washout (clearance), which in the spinotrapezius muscle is reported to be a simple mono-exponential process (Hyman & Paldino, 1962; Hyman et al. 1963). It is important to note that both weakly and strongly fluorescent DAF-FM forms undergo washout from the interstitium, which complicates the prediction of the time course of fluorescence during a bout of muscle contraction. This is an important effect undermining the measurement of the physiological concentration of NO in tissues in situ (Hall & Garthwaite, 2009). An additional challenge is created by the change in indicator washout rate from active hyperaemia caused by contractions of the stimulated muscle. To account for this effect, a registration of the clearance curve before and after the bout of contraction is desirable.

To predict the time course of fluorescence in response to a series of muscle contractions in a previously resting muscle, we assume that the onset of muscular contraction withdraws superoxide from the interstitial space and makes NO available for reaction with DAF-FM. This means that the response of interstitial DAF-FM to a step function of the concentration of NO has to be analysed.

Conversion of DAF-FM into the highly fluorescent form due to interaction with NO is known to be a multistage process with different rates of reactions (Namin et al. 2013). Assuming that the overall rate is determined by the slowest reaction, we take simplified first order kinetics for the conversion of DAF-FM into DAF-T. We also assume that the concentration of NO in the interstitial space does not limit the rate of reaction, due to the permanent production of NO by the endothelium. This approach provides an opportunity to formulate a conceptual description of the fluorescence time course caused by muscle contraction, combined with the clearance of both forms of the indicator. Photo-bleaching is not considered, so an important condition for the proposed experiment is that photo-bleaching of the DAF-T is minimized to negligible levels.

In this basic model, the concentration of weakly fluorescent DAF-FM in the interstitial fluid decreases due to washout and to conversion to DAF-T. At the same time, the concentration of highly fluorescent DAF-T increases due to conversion from DAF-FM, but decreases due to clearance. This can be expressed in two equations:

graphic file with name tjp0592-1225-m1.jpg (1)
graphic file with name tjp0592-1225-m2.jpg (2)

where t is time from the beginning of the process, u is [DAF-FM], y is [DAF-T], k is the rate of conversion of DAF-FM to DAF-T, f is the rate of washout of the DAF-FM and DAF-T indicators. The solution of this system of equations is:

graphic file with name tjp0592-1225-m3.jpg (3)

where u0 is [DAF-FM] and y0 is [DAF-T] at time t = t0, corresponding to the moment of onset of contractions in a previously resting muscle. In the absence of probe reaction with NO (k = 0) the equation describes the exponential curve of indicator washout. In the absence of clearance of indicators (f = 0) it describes the rising exponential curve of fluorescence with asymptote (y0 + u0). Another form of this equation shows the role of the ratio y0/u0 at the initial moment of the process, and that of the coefficients which determine the reaction rate and clearance (k and f, respectively) to the overall shape of the fluorescent signal:

graphic file with name tjp0592-1225-m4.jpg (4)

Two families of curves generated by eqn (4), for example at y0/u0 = 0.1, k = 0.005 s−1, f = 0.005 s−1 and for variable clearance and reaction rates, are shown in Fig. 2. This implies that getting a strong fluorescent signal requires not only a high reaction rate and a slow clearance, but also a low initial ratio y0/u0. The latter fact imposes a limitation on the timing of loading the probe into the interstitium of the spinotrapezius muscle and on the timing of the experimental protocol. The penetration of the probe throughout a thin spinotrapezius muscle takes about half an hour, but during that time there is a slow conversion of the probe into the fluorescent form, thus increasing y0/u0. The duration of the experiment is also limited by washout of the probe from the muscle which decreases the signal-to-noise ratio in fluorescence measurements.

Figure 2.

Figure 2

Open in a new tab

See eqns (3) and (4). In the left panel the apparent rate of reaction of DAF-FM with NO is fixed and washout rates are varied. In the right panel the washout rate is fixed and reaction rates are variable. The 5-to 10-fold predominance of the washout rate over the reaction rate makes the fluorescence signal not detectable.

The ratio of the rate of DAF-FM to DAF-T conversion and the rate of probe washout is crucial for the time course of the fluorescence rise caused by a bout of muscle contraction (Fig. 2). If the probe is washed out 5 times faster compared to its conversion rate, the fluorescence excess can hardly be detected against the background process of DAF-T washout. Because we did not intend to eliminate or slow down the clearance of the indicator from the interstitium, we accounted for it in a model of the fluorescence curve and expected that the similarity of the experimental and predicted curves can testify to the correctness of the assumptions underlying the model.

Stimulated muscle contraction is known as the cause of active hyperaemia, which changes the clearance rate for both forms of the indicator (Hyman & Paldino, 1962; Hyman et al. 1963). It is difficult to establish the increase of the clearance rate during a period of muscle stimulation; therefore it is useful to measure the clearance rates f1 and f2, before and after the round of contractions, respectively. For ease of operations with the experimental data in the text, instead of very small fractions f1 and f2 we also use the inverse values in presentation of the experimentally obtained parameters: F1 = 1/f1; F2 = 1/f2.

Experimental animals

All procedures were approved by the Virginia Commonwealth University Institutional Animal Care and Use Committee and are consistent with the National Institutes of Health guidelines for the humane treatment of laboratory animals, the American Physiological Society's Guiding Principles in the Care and Use of Animals and conform to the principles of UK regulations, as described by Drummond (2009).

Male Sprague-Dawley rats (Harlan, Indianapolis, IN, USA) were initially anaesthetized with an intraperitoneal injection of a ketamine and acepromazine mixture (72 mg kg−1 ketamine and 3 mg kg−1 acepromazine). The trachea was cannulated with PE-240 tubing to maintain a patent airway. The jugular vein was cannulated with PE 90 tubing for continuous intravenous infusion of alfaxalone (Alfaxan, Schering-Plough Animal Health, Welwyn Garden City, UK; approximately 0.1 mg kg−1 min−1). The anaesthetic rate was adjusted as needed to maintain a surgical plane of anaesthesia, as evidenced by the absence of a toe pinch reflex and stable vital signs. Heart rate and blood oxygen saturation were constantly monitored by a pulse oximeter (PulseSense, Nonin Medical, Plymouth, MN, USA). Body and spinotrapezius muscle temperatures were maintained at 37°C by a thermostatic animal platform (Golub & Pittman, 2003). Experiments were carried out on 11 rats with average body weight of 376 ± 6 g, average heart rate of 271 ± 8 min−1 and oxygen saturation of 93 ± 1%.

Animals were killed at the end of experimentation by giving the anaesthetized rats an overdose of Euthasol (0.4 ml kg−1, i.v.; consisting of 390 mg ml−1 sodium pentobarbital and 50 mg ml−1 phenytoin mixture) at the end of the experiment. This method is consistent with the recommendations of the Panel on Euthanasia of the American Veterinary Medical Association.

Surgical preparation of the spinotrapezius muscle was similar to the original description (Gray, 1973; Bailey et al. 2000). The muscle was gently spread on a thermo-stabilized (37°C) pedestal of the animal platform and fixed with sutures to a rigid frame, to ensure isometric contractions. The muscle was covered with a gas barrier film (Krehalon, Kureha, Japan). An objective-mounted film airbag connected to a pressure controller allowed permanent organ compression at 5 mmHg, which ensured a tight seal of the film to the muscle and stabilized its position during contraction (Golub et al. 2011). Two thin chlorinated silver wire electrodes were placed on the sides of the muscle for electrical stimulation of muscle contractions (1 Hz, 10 V, 20 ms).

Microscopic pulse fluorimetry

The pulse fluorescence instrument consisted of an Axio Imager A2m microscope (Zeiss, Germany), a photomultiplier unit (R9110 and C7950, Hamamatsu Corp., Bridgewater, NJ, USA) and a video camera (KP-D20BU; Hitachi, Japan). Epi-illumination excitation was provided by a 505 nm (cyan) light emission diode (LXHL-LE5C, http://www.luxeon.com) with a strobe driver (SD-1000, http://www.stockeryale.com) that generated 500 μs rectangular light pulses at a rate of 1 Hz and current 1 A. The light beam inside the microscope was directed by a fluorescence filter cube (Blue 11001; Chroma Technology Corp., Rockingham, VT, USA) through a ×20, 0.8 NA objective lens onto the tissue, where it formed an excitation spot 1 mm in diameter (detection area on the photomultiplier was limited to 0.78 mm). Fluorescence emission was detected by a photomultiplier and the signal was digitized at a sampling rate of 100 kHz with a 12-bit analog-to-digital converter (Model PC-MIO-16E-4; National Instruments, Austin, TX, USA). Data acquisition and analysis of signals were performed with custom codes using LabVIEW (National Instruments).

Experimental protocol

The fluorescent probe DAF-FM (http://www.invitrogen.com) was dissolved with phosphate buffered saline to a concentration of 0.5 mm and loaded into the interstitial space via topical application to the surface of the spinotrapezius muscle for 20–30 min. Circular regions of muscle of radius 0.78 mm containing no large microvessels were selected for DAF-T fluorescence measurements.

The fluorescence intensity was sampled once a second, synchronously with the electrical stimulator pulse, so that 50 data points (0.5 ms) were recorded just before the next muscle contraction began, which occurred more than 20 ms later due to electromechanical delay. The full record of the fluorescence signal included a time interval of resting muscle (baseline), a bout of rhythmic contraction and a post-contraction hyperaemic period. During the first 2–5 min the fluorescence decrease in the interstitial space of the resting muscle was recorded. Then the muscle was electrically stimulated for 3–5 min and, after the bout of contraction, the fluorescence in the resting muscle was recorded for an additional 3–5 min. The variability of the rest and stimulation periods was considered in order to fit the signal within the upper input range (10 V) of the analog-to-digital converter. It took several minutes to move the detection area to a new site and measurements were repeated in a different region of the muscle.

Tests

The short-pulse fluorescence measurement technique was used to reduce the effect of photo-bleaching. This effect was tested using sodium fluorescein, instead of the weakly fluorescent DAF-FM (and unavailable DAF-T), and photo-bleaching was found to be very low (0.8% per 1000 light pulses in a 1 mm solution). In order to evaluate the possible effects of changes in muscle thickness on the amplitude of the fluorescence signal during isometric contractions, the fluorescence signals and a short light pulse from a trans-illuminating green light emission diode were simultaneously recorded. During this test the muscle was stimulated at 1 Hz and the light pulse illuminated at 100 Hz, an arrangement which provided good time resolution for each contraction. A small synchronous change in fluorescence amplitude (12%) and light transmission (21%) were detected for a time interval of 200 ms, but both parameters were totally restored to previous levels before the moment of the next fluorescence measurement.

To avoid the influence of a specific reaction of DAF-FM with tissue NO on the time course of fluorescence in skeletal muscle, sodium fluorescein was used for the indicator washout test. This inert indicator has about the same molecular weight as DAF-FM (molecular weights are 376 and 412, respectively) and is also distributed only in the interstitial space. The test demonstrated a mono-exponential decrease of the fluorescence intensity of these indicators, which explains this effect for DAF-FM by washout of low molecular weight substances from the interstitial space.

A curve fitting procedure was used to recover parameters from the experimental curves. A mono-exponential model was used to determine the time constants of the indicator washout during the resting states of the muscle, before and after the bout of contractions. The Levenberg-Marquardt algorithm was applied to fit the fluorescence time course during the bout of muscle contraction, using a modified form of eqn (4) as a fitting model:

graphic file with name tjp0592-1225-m5.jpg (5)

where t0 is the moment of the onset of 1 Hz contractions.

Statistical calculations and parameter fitting were made with the OriginPro 8.1 (OriginLab Corp., Northampton, MA, USA) software package. All data are presented as means ± SEM (number of measurements).

Results

In resting spinotrapezius muscle the interstitial concentration of sodium fluorescein decreased according to a mono-exponential time course. As the photo-bleaching was very small, the exponential decrease in fluorescence of this inert indicator was determined only by washout rate. The time constants were: F1 = 430 ± 74 s (n = 7) at rest, Fst = 380 ± 98 s (n = 4) during the 200 s period of stimulated contractions and F2 = 128 ± 29 s (n = 3) for the post-contraction period. Comparison of the time constants of the indicator washout before and after the bout of muscle contraction, made in the total set of experimental curves (n = 49), showed a statistically significant (at the P = 0.01 level) increase of washout rate in the post-contraction period from the time constant at rest F1 = 289 ± 34 s (n = 49) to F2 = 154 ± 23 s (n = 44) after the end of stimulation. These results are consistent with long-known data on the increased washout rate following muscle contraction (Hyman & Paldino, 1962; Hyman et al. 1963). Washout of both the DAF-FM and DAF-T forms of the probe from the interstitial space was one of the reasons that impaired the quality of the signal in repeated cycles of measurement. Another reason for the reduced intensity of the fluorescent signal from one cycle to another was presumably related to the development of sustainable hyperaemia and fatigue of the muscle preparation.

For the analysis of fluorescence time course associated with a bout of muscle contraction, 32 curves having the salient fluorescence signal protruding above the curve of indicator washout were selected. Examples of these experimental curves illustrating the effect of various combinations between the reaction and washout rates in the muscle areas are shown in Fig. 3. The initial segment of each recorded curve corresponds to the resting state and the mono-exponential washout of indicator is the dominant process. The onset of muscle contraction caused a sharp ascent of fluorescence intensity, while the cessation of stimulation immediately led to a mono-exponential washout curve, which was usually faster than that before the bout of contractions (Fig. 4). Despite the variability of forms of different fluorescence curves during the period of muscle contraction, they were all well fitted with eqn (5) (see the family of curves in Fig. 2A and B), which allowed us to estimate the values of the fitting parameters.

Figure 3.

Figure 3

Open in a new tab

The horizontal line indicates the period of muscle contractions every second. Parameters of example curves expressed in units of (s−1): f1 = 0.005, f2 = 0.013, k = 0.032 in curve A; f1 = 0.006, f2 = 0.007, k = 0.001 in curve B; f1 = 0.007, f2 = 0.008, k = 0.002 in curve C; f1 = 0.004, f2 = 0.006, k = 0.062 in curve D.

Figure 4.

Figure 4

Open in a new tab

k = 0.01 ± 0.003 s−1; f1 = 0.009 ± 0.002 s−1 and f2 = 0.012 ± 0.002 s−1. This display shows the mono-exponential decrease of fluorescence due to washout of the indicator during 200 s at rest, followed by a 200 s bout of muscle contractions at 1 Hz and then a 200 s time period of rest and washout of the indicator. The ratio of the reaction rate of DAF-FM with NO and rate of indicator washout allows reliable detection of the fluorescence peak associated with the increased availability of NO during muscle contraction.

Estimated by the curve fitting and expressed in arbitrary units of fluorescence intensity (AU), the initial concentrations of both forms of DAF were y0 = 4.6 ± 0.2 (n = 32) and u0 = 129 ± 42 (n = 32). Thus, the ratio y0/u0, indicating the fraction of converted form DAF-T at the onset of contraction, was 0.036, even lower than that expected (y0/u0 = 0.1) in the simulation using eqn (4) and shown in Fig. 2. This means that only a small amount of DAF-FM was converted during loading the probe and it provides a good starting condition for recording the fluorescent response to muscle contraction. The average conversion rate of DAF-FM into DAF-T, k = 0.01 ± 0.003 s−1 (n = 31), is sufficiently high to produce a detectable fluorescence peak pattern according to Fig. 2 (right panel). The washout rate during contraction was in the range between the value at rest f1 = 0.009 ± 0.002 s−1 (n = 26) and that after contraction f2 = 0.012 ± 0.002 s−1 (n = 30). As follows from Fig. 2 (left panel), at a washout rate of about 0.01 s−1 there is a strong suppression of the fluorescence peak, so that the possibility exists for a very poor response at a high indicator washout rate in the detection area. Thus, the various curves of DAF-T fluorescence in contracting muscle, as shown in Fig. 3, are generated by the same process which is well described by the model (eqns (1) and (2), Fig. 2), but look different due to variations in the balance of rates of DAF-T production and washout.

The theoretical curve constructed by substituting the average values of obtained parameters into eqn (4) is shown in Fig. 4. The pre-and post-contraction periods of the muscle at rest are represented by mono-exponential washout curves (k ≈ 0), while the period of stimulated muscle contraction was calculated using a simulation of the DAF-FM response to a step increase in the concentration of NO (eqn (4). This curve reflects the general features of a set of experimental curves: the mono-exponential clearance of the indicators (DAF-FM and DAF-T) before the 200 s period of muscle contraction, an immediate sharp increase in fluorescence intensity after the onset of contraction, followed by a slower growth and decline due to a limitation of DAF-T production and washout of both forms of the indicator, and finally the increased washout rate due to active hyperaemia after the contraction period.

Discussion

The proposed model of local blood flow regulation departs from the traditional metabolic hypothesis (Rowell, 2004). Exercise is considered to be the normal physiological state of a skeletal muscle and the anatomical dimension of the associated vascular bed is adapted to the maximal metabolic rate of the muscle (Weibel & Hoppeler, 2005). Thus, vasodilatation is considered to be the normal state of the arterioles (working as a normally open gate valve in local blood flow regulation). The regulatory action on the vasculature is vasoconstriction aimed at reducing local blood flow to prevent excessive oxygen and glucose supply to parenchymal cells. Instead of seeing the nitric oxide and superoxide radicals as toxic byproducts of metabolism and damaging agents, this hypothesis considers them as a complementary pair of directly interacting messengers that provide rapid signalling between the vasculature and musculature, thus abolishing the need to search for a hypothetical metabolic vasodilator. This feedback control system is built from well-studied components: the radical generating enzymes (eNOS and NAD(P)H oxidase) and a complementary pair of oxygen-linked radical messengers (NO and O2−). This regulatory mechanism links the rate of tissue perfusion to the activity of mitochondria and explains the behaviour of the system in response to changes in metabolic rate.

Extracellular superoxide dismutase (ecSOD) can play an important role in the regulation scheme because it accelerates the withdrawal of the braking signal carried by the interstitial flux of superoxide (Oury et al. 1996; Demchenko et al. 2010), contributing to the rapid growth of the interstitial concentration of NO and the fluorescence of the indicator. However, the peculiarities of ecSOD in rats (Karlsson & Marklund, 1988) call into question the role of this enzyme in the local balance of NO and superoxide. In contrast to the tetrameric ecSOD in other species, this enzyme is a low molecular weight (85 kD) dimer and it does not possess a heparin-binding domain (Karlsson & Marklund, 1988; Carlsson et al. 1996). These properties determine its high turnover rate and low concentration in the interstitium of rat muscle, while its activity in plasma is two orders of magnitude higher than that in human (Karlsson & Marklund, 1987, 1988). These peculiarities should be taken into account for the translation of experimental results to the human circulation. Since the content of ecSOD in the perivascular interstitium in humans is very large, the role of this enzyme in humans should be far more important than that in rats (Stralin et al. 1995; Golub & Pittman, 2013).

We also recognize the possible involvement of NAD(P)H oxidase, located in the vascular wall, and nNOS, located in muscular cells, in this model of regulation, but did not consider them now for the sake of simplicity. For this reason, we leave aside the discussion of well-known external activators for the key enzymes in the proposed model: wall shear stress for eNOS and the angiotensin II/AT1 pathway for NAD(P)H oxidase. From the general scheme it is clear that their participation does not contradict the basic principles of operation of the regulator (Fig. 1).

A feature of the proposed regulation circuit is the presentation of the reference and feedback signals in the form of production rates of interacting radicals (Fig. 1). Since the equality of rates is unlikely, the change in rates near their equilibrium values must rapidly switch the error signal, [NO], between its high and low values. This means that the negative feedback regulator may not be a proportional type, but rather a bang–bang (or on–off) controller, characterized by a threshold level and a dead band. At the threshold level the error signal switches to its low state and it remains insensitive within the dead band range of input signals, thus providing a hysteresis type of response. Regulators of this type are characterized by simplicity of design and high speed of control responses. Examples of such controls are a float regulator of fluid level, a home thermostat, and so on. The main feature of this type of controller is its bistability, that is, the existence of two stable states. The argument in favour of bistability of local blood flow control is the existence of two additional (not presented in the diagram) positive feedback loops that facilitate switching of the error signal between the high and low levels: (1) increase of blood flow in the microvessels stimulates the production of NO by the endothelium via a wall shear stress mechanism (Koller & Kaley, 1990a; Koller & Bagi, 2002; Toth et al. 2007; Andrews et al. 2011) and (2) the excess supply of oxygen and glucose to parenchymal cells increases the production of superoxide by membrane NAD(P)H oxidase (Javesghani et al. 2002; Chen et al. 2005; Afanas'ev, 2010).

The results of the present experiments are consistent with the predicted increase of signal from the fluorescent probe DAF-FM, distributed in the interstitial fluid, due to its transformation into a highly fluorescent form DAF-T during a bout of muscle contraction. Moreover, the fluorescence time course is well described by an equation derived from a model of DAF-FM reaction with a rectangular pulse of NO, combined with a washout of both forms of the indicator from the interstitium (eqn (4), Fig. 2). Equation (4) was also used for fitting the parameters of the experimental curves before and after the bout of muscle contraction (k = 0). Earlier measurements of the washout of the indicator in rat spinotrapezius muscle (Water Blue; Hyman & Paldino, 1962; Hyman et al. 1963), performed under superfusion conditions in the presence of atmospheric oxygen, determined a clearance rate of 0.0533 min−1 (the corresponding time constant was F1 = 1126 s). In our experiments the clearance of the indicator was comparatively faster during the period of rest before a bout of contraction (F1 = 289 s). We attribute this difference to isolation of the tissue from atmospheric oxygen in our preparations and also to residual hyperaemia developed after the series of previous muscle stimulations. It is well known that the clearance rate is associated with local blood flow. We observed an almost 2-fold (time constant change from 289 to 154 s) acceleration of the clearance rate following a bout of muscle contraction, confirming the development of post-contraction or active hyperaemia.

Thus, the washout of an indicator rapidly worsens the quality of the fluorescent signal, a problem which must be considered in studies using interstitially distributed indicators (Fig. 2). However, 32 of the 42 curves obtained had a good signal and a prominent fluorescent peak suitable for curve fitting analysis with eqn (5). A representative curve (Fig. 4) constructed from the average parameters using eqn (4) showed a clear and immediate increase of fluorescence after the start of muscle contractions. A gallery of examples of experimental curves (Fig. 3) demonstrates a diversity of curve shapes and the immediate increase in fluorescence associated with the onset of contractions, with no smooth transitions from rest to exercise periods. All these curves are well described by the same equation simulating the reaction of indicator with NO together with the background clearance (eqn (4); Fig. 2). The fluorescence curves show that the reaction rate of DAF-FM with NO is high enough to detect the increase of NO availability, despite having the background process of indicator washout. These experimental results support the hypothesis that a sharp increase in muscle metabolism curtails the neutralization of interstitial NO with superoxide, which drives the well-expressed active hyperaemia (Golub & Pittman, 2013).

Moreover, these results support the hypothesis of the bang–bang mechanism of regulation, because the fluorescent signal has a rapid rise at the onset of contraction and a rapid transition to a mono-exponential clearance at its end (Fig. 4). This behaviour is consistent with an immediate rise in muscle respiration rate, which, however, requires approximately one minute to achieve a steady state level, and differs from the response of microvascular Inline graphic, which begins to decline after a delay of about 20 s (Poole et al. 2011). Restoration of microvascular Inline graphic following muscle contraction also is delayed. These delays, found in an identical experimental protocol, do not support the assumption of increased fluorescence due to stimulated production of NO by the endothelium, since this process would have delays after onset and cessation of contractions. However, the flow-stimulated production of NO (Koller & Kaley, 1990a,1990b; Koller & Bagi, 2002) is consistent with the proposed model (Fig. 1).

It must be emphasized that in this model the regulatory response depends on the delivery of both oxygen and glucose. The intensity of glycolysis determines the turnover of cytosolic NADH. A high rate of glycolysis provides a high yield of NADH and NADPH into the sub-sarcolemmal cytosol (Wolin et al. 1999; Didion & Faraci, 2002). These cofactors are substrates for membrane NAD(P)H oxidase, which transforms interstitial oxygen to superoxide depending on the O2 concentration in the interstitium. In addition to the direct substrate dependence for NAD(P)H oxidase, there are several known signalling pathways linking glucose with the increased production of superoxide (Afanas'ev, 2010). The strong oxygen dependence of NAD(P)H oxidase has been established experimentally, with a Km of about 12–15 mmHg and a maximal production rate at a Inline graphic of about 80 mmHg (Archer et al. 1993; Basini et al. 2004; Chen et al. 2005). Thus, a decrease in oxygen concentration leads to a reduction of the superoxide neutralizing effect on the interstitial concentration of NO and to vasodilatation, giving the appearance of a vascular response to hypoxia.

We have discussed the hypothetical principle of local blood flow regulation applied particularly to the microcirculation in skeletal muscle; however, we suggest a wider applicability of this model to most organs. Experimental data have been accumulated in the literature about the key role of intercellular signalling via NO–O2− interaction, especially in the regulation of neurovascular coupling. Measurement of NO with a microelectrode placed in the somatosensory cortex of rats revealed a peak transient of [NO] in response to forepaw stimulation (Buerk et al. 2003). Following the peak of [NO], local blood flow increased, while the intensity of the [NO] signal decreased. This behaviour is similar to our results obtained in muscle, adjusted for the fact that the fluorescent probe integrates the [NO] signal. Long-term studies of the mechanisms causing vasoconstriction in the brain in response to hyperbaric oxygen have revealed the central role of superoxide in the inactivation of NO, and participation of superoxide dismutase in the control of the NO–O2− balance (Demchenko et al. 2000a b2000b, 2002, 2010; Zhilyaev et al. 2003). Furthermore, the rapidity and precise location of local blood flow responses in the brain, enabling the interpretation of fMRI signals, are due in large part to the presence of extracellular superoxide dismutase in the brain; this is not as important for muscle. Considered together, these reports support the suggestion that local functional hyperaemia in the brain is regulated by a mechanism similar to that proposed for skeletal muscle.

Conclusion

In a proof-of-concept experiment we have tested the hypothesis that the regulation of local blood flow is realized through neutralizing nitric oxide, steadily produced by microvascular eNOS, with its antagonist, superoxide anion, generated into the interstitial space by sarcolemmal NAD(P)H oxidase. This hypothetical regulatory mechanism is built from well-studied components: the radical generating enzymes, eNOS and NAD(P)H oxidase, and the resulting complementary pair of radical messengers, NO and O2−, and the O2− scavenging enzyme ecSOD. In contrast to the dominant metabolic hypothesis of local blood flow regulation, the new model does not contain unknown elements and, thus, can be experimentally verified or falsified. The neutralization of interstitial NO takes place under conditions of an adequate or excessive supply of oxygen and glucose to the parenchymal cells, which leads to the production of sufficient amounts of cytosolic NADH and NADPH, and generation of superoxide by sarcolemmal NAD(P)H oxidase into the interstitium, where it actively reacts with NO. Thus, under the conditions of an adequate or excessive supply of oxygen and glucose to tissue cells, the interstitial concentration of NO is low. In this case, the onset of muscle contraction quickly reduces the production of superoxide, leading to an increase in the concentration of NO in the interstitium. This effect can be detected by its reaction with the fluorescent probe DAF-FM. In in situ experiments the peaks of fluorescence intensity in response to a bout of muscle contraction were detected and their shape was distorted by the high rate of indicator washout. The shape of the transients in DAF-T fluorescence corresponds to predictions of a mathematical model which was used for parameter fitting of experimental curves. The rapid time course of the fluorescent signal during the transition from rest to exercise and back indicates the possibility of an on–off type of regulation. In our opinion, this hypothetical model for the regulation of local blood flow can explain physiological and patho-physiological processes in the microcirculation without the involvement of elusive metabolic vasodilators. We believe that this mechanism of regulation, studied in this work in skeletal muscle, is common to most organs and tissues of mammals.

Key points

  • The metabolic theory of blood flow regulation suggests that, when tissue cells experience a reduction of oxygen supply, they produce metabolic vasodilators, which increase the lumen of arterioles and hence local blood flow.

  • A century of intensive research has not found a single metabolic vasodilator to account for observed flow changes, therefore current thought is that there are many vasodilators.

  • We have proposed an alternative hypothesis based on the interaction of two well-known molecular mechanisms for generation and removal of the intercellular signalling radicals nitric oxide (NO) and superoxide.

  • The proposed mechanism of regulation predicts a sharp increase in NO concentration in the intercellular space at the onset of muscle contraction.

  • Experiments with the NO-sensitive fluorescent indicator DAF-FM, loaded into the intercellular space, confirmed the rapid response of the NO-related signal at the beginning of contractions and rapid washout of the indicator after their termination.

Glossary

DAF-FM

4-amino-5-methylamino-2′,7′-difluorofluorescein

DAF-T

triazole derivative of DAF-FM

ecSOD

extracellular superoxide dismutase

eNOS

endothelial nitric oxide synthase

NADH

nicotinamide adenine dinucleotide

NADPH

nicotinamide adenine dinucleotide phosphate

NO

nitric oxide

O2−

superoxide anion

SMC

smooth muscle cell

Additional information

Competing interests

The authors declare no conflict of interest.

Author contributions

A.S.G.: conception and design of experiment; collection, analysis and interpretation of the data and drafting of the manuscript. B.K.S.: design of experiment, collection and analysis of the data. R.N.P.: conception and design of experiment; analysis and interpretation of the data and drafting of the manuscript. All authors read and approved the final version of the manuscript.

Funding

This study was supported by grant HL18292 from the National Heart, Lung and Blood Institute, by funding from the VCU Presidential Research Quest Fund and by research support funds from the VCU School of Medicine.

References

  1. Afanas'ev I. Signaling of reactive oxygen and nitrogen species in Diabetes mellitus. Oxid Med Cell Longev. 2010;3:361–373. doi: 10.4161/oxim.3.6.14415. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Andrews AM, Jaron D, Buerk DG, Kirby PL, Barbee KA. Direct, real-time measurement of shear stress-induced nitric oxide produced from endothelial cells in vitro. Nitric Oxide. 2011;23:335–342. doi: 10.1016/j.niox.2010.08.003. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Archer SL, Huang J, Henry T, Peterson D, Weir EK. A redox-based O2 sensor in rat pulmonary vasculature. Circ Res. 1993;73:1100–1112. doi: 10.1161/01.res.73.6.1100. [DOI] [PubMed] [Google Scholar]
  4. Bailey JK, Kindig CA, Behnke BJ, Musch TI, Schmid-Schoenbein GW, Poole DC. Spinotrapezius muscle microcirculatory function: effects of surgical exteriorization. Am J Physiol Heart Circ Physiol. 2000;279:H3131–H3137. doi: 10.1152/ajpheart.2000.279.6.H3131. [DOI] [PubMed] [Google Scholar]
  5. Basini G, Grasselli F, Bianco F, Tirelli M, Tamanini C. Effect of reduced oxygen tension on reactive oxygen species production and activity of antioxidant enzymes in swine granulosa cells. Biofactors. 2004;20:61–69. doi: 10.1002/biof.5520200201. [DOI] [PubMed] [Google Scholar]
  6. Beckman JS, Koppenol WH. Nitric oxide, superoxide, and peroxynitrite: the good, the bad, and ugly. Am J Physiol Cell Physiol. 1996;271:C1424–C1437. doi: 10.1152/ajpcell.1996.271.5.C1424. [DOI] [PubMed] [Google Scholar]
  7. Buerk DG, Ances BM, Greenberg JH, Detre JA. Temporal dynamics of brain tissue nitric oxide during functional forepaw stimulation in rats. Neuroimage. 2003;18:1–9. doi: 10.1006/nimg.2002.1314. [DOI] [PubMed] [Google Scholar]
  8. Carlsson LM, Marklund SL, Edlund T. The rat extracellular superoxide dismutase dimer is converted to a tetramer by the exchange of a single amino acid. Proc Natl Acad Sci U S A. 1996;93:5219–5222. doi: 10.1073/pnas.93.11.5219. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Chen Y, Gill PS, Welch WJ. Oxygen availability limits renal NADPH-dependent superoxide production. Am J Physiol Renal Physiol. 2005;289:F749–F753. doi: 10.1152/ajprenal.00115.2005. [DOI] [PubMed] [Google Scholar]
  10. Demchenko IT, Boso AE, Bennett PB, Whorton AR, Piantadosi CA. Hyperbaric oxygen reduces cerebral blood flow by inactivating nitric oxide. Nitric Oxide. 2000a;4:597–608. doi: 10.1006/niox.2000.0313. [DOI] [PubMed] [Google Scholar]
  11. Demchenko IT, Boso AE, O'Neill TJ, Bennett PB, Piantadosi CA. Nitric oxide and cerebral blood flow responses to hyperbaric oxygen. J Appl Physiol. 2000b;88:1381–1389. doi: 10.1152/jappl.2000.88.4.1381. [DOI] [PubMed] [Google Scholar]
  12. Demchenko IT, Gutsaeva DR, Moskvin AN, Zhilyaev SY. Involvement of extracellular superoxide dismutase in regulating brain blood flow. Neurosci Behav Physiol. 2010;40:173–178. doi: 10.1007/s11055-009-9240-5. [DOI] [PubMed] [Google Scholar]
  13. Demchenko IT, Oury TD, Crapo JD, Piantadosi CA. Regulation of the brain's vascular responses to oxygen. Circ Res. 2002;91:1031–1037. doi: 10.1161/01.res.0000043500.03647.81. [DOI] [PubMed] [Google Scholar]
  14. Didion SP, Faraci FM. Effects of NADH and NADPH on superoxide levels and cerebral vascular tone. Am J Physiol Heart Circ Physiol. 2002;282:H688–H695. doi: 10.1152/ajpheart.00576.2001. [DOI] [PubMed] [Google Scholar]
  15. Drummond GB. Reporting ethical matters in The Journal of Physiology: standards and advice. J Physiol. 2009;587:713–719. doi: 10.1113/jphysiol.2008.167387. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Figueroa XF, Martinez AD, Gonzalez DR, Jara PI, Ayala S, Boric MP. In vivo assessment of microvascular nitric oxide production and its relation with blood flow. Am J Physiol Heart Circ Physiol. 2001;280:H1222–H1231. doi: 10.1152/ajpheart.2001.280.3.H1222. [DOI] [PubMed] [Google Scholar]
  17. Golub AS, Pittman RN. Thermostatic animal platform for intravital microscopy of thin tissues. Microvasc Res. 2003;66:213–217. doi: 10.1016/s0026-2862(03)00061-x. [DOI] [PubMed] [Google Scholar]
  18. Golub AS, Pittman RN. Bang-bang model for regulation of local blood flow. Microcirculation. 2013;20:455–483. doi: 10.1111/micc.12051. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Golub AS, Tevald MA, Pittman RN. Phosphorescence quenching microrespirometry of skeletal muscle in situ. Am J Physiol Heart Circ Physiol. 2011;300:H135–H143. doi: 10.1152/ajpheart.00626.2010. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Gray SD. Rat spinotrapezius muscle preparation for microscopic observation of the terminal vascular bed. Microvasc Res. 1973;5:395–400. doi: 10.1016/0026-2862(73)90055-1. [DOI] [PubMed] [Google Scholar]
  21. Hall CN, Garthwaite J. What is the real physiological NO concentration in vivo. Nitric Oxide. 2009;21:92–103. doi: 10.1016/j.niox.2009.07.002. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Hyman C, Paldino RL. Local temperature regulation of microtissue clearance from rat skeletal muscle. Circ Res. 1962;10:89–93. doi: 10.1161/01.res.10.1.89. [DOI] [PubMed] [Google Scholar]
  23. Hyman C, Paldino RL, Zimmermann E. Local regulation of effective blood flow in muscle. Circ Res. 1963;12:176–181. doi: 10.1161/01.res.12.2.176. [DOI] [PubMed] [Google Scholar]
  24. Javesghani D, Magder SA, Barreiro E, Quinn MT, Hussain SN. Molecular characterization of a superoxide-generating NAD(P)H oxidase in the ventilatory muscles. Am J Respir Crit Care Med. 2002;165:412–418. doi: 10.1164/ajrccm.165.3.2103028. [DOI] [PubMed] [Google Scholar]
  25. Karlsson K, Marklund SL. Heparin-induced release of extracellular superoxide dismutase to human blood plasma. Biochem J. 1987;242:55–59. doi: 10.1042/bj2420055. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Karlsson K, Marklund SL. Extracellular superoxide dismutase in the vascular system of mammals. Biochem J. 1988;255:223–228. [PMC free article] [PubMed] [Google Scholar]
  27. Kissner R, Nauser T, Bugnon P, Lye PG, Koppenol WH. Formation and properties of peroxynitrite as studied by laser flash photolysis, high-pressure stopped-flow technique, and pulse radiolysis. Chem Res Toxicol. 1997;10:1285–1292. doi: 10.1021/tx970160x. [DOI] [PubMed] [Google Scholar]
  28. Koller A, Bagi Z. On the role of mechanosensitive mechanisms eliciting reactive hyperemia. Am J Physiol Heart Circ Physiol. 2002;283:H2250–H2259. doi: 10.1152/ajpheart.00545.2002. [DOI] [PubMed] [Google Scholar]
  29. Koller A, Kaley G. Endothelium regulates skeletal muscle microcirculation by a blood flow velocity-sensing mechanism. Am J Physiol Heart Circ Physiol. 1990a;258:H916–H920. doi: 10.1152/ajpheart.1990.258.3.H916. [DOI] [PubMed] [Google Scholar]
  30. Koller A, Kaley G. Role of endothelium in reactive dilation of skeletal muscle arterioles. Am J Physiol Heart Circ Physiol. 1990b;259:H1313–H1316. doi: 10.1152/ajpheart.1990.259.5.H1313. [DOI] [PubMed] [Google Scholar]
  31. Nagano T. Bioimaging probes for reactive oxygen species and reactive nitrogen species. J Clin Biochem Nutr. 2009;45:111–124. doi: 10.3164/jcbn.R09-66. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Namin SM, Nofallah S, Joshi MS, Kavallieratos K, Tsoukias NM. Kinetic analysis of DAF-FM activation by NO: toward calibration of a NO-sensitive fluorescent dye. Nitric Oxide. 2013;28:39–46. doi: 10.1016/j.niox.2012.10.001. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Oury TD, Day BJ, Crapo JD. Extracellular superoxide dismutase: a regulator of nitric oxide bioavailability. Lab Invest. 1996;75:617–636. [PubMed] [Google Scholar]
  34. Poole DC, Copp SW, Hirai DM, Musch TI. Dynamics of muscle microcirculatory and blood-myocyte O2 flux during contractions. Acta Physiol (Oxf) 2011;202:293–310. doi: 10.1111/j.1748-1716.2010.02246.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Rowell LB. Ideas about control of skeletal and cardiac muscle blood flow (1876–2003): cycles of revision and new vision. J Appl Physiol. 2004;97:384–392. doi: 10.1152/japplphysiol.01220.2003. [DOI] [PubMed] [Google Scholar]
  36. Roy CS, Brown JG. The blood-pressure and its variations in the arterioles, capillaries and smaller veins. J Physiol. 1880;2:323–359. doi: 10.1113/jphysiol.1880.sp000068. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Stralin P, Karlsson K, Johansson BO, Marklund SL. The interstitium of the human arterial wall contains very large amounts of extracellular superoxide dismutase. Arterioscler Thromb Vasc Biol. 1995;15:2032–2036. doi: 10.1161/01.atv.15.11.2032. [DOI] [PubMed] [Google Scholar]
  38. Stuehr DJ, Santolini J, Wang ZQ, Wei CC, Adak S. Update on mechanism and catalytic regulation in the NO synthases. J Biol Chem. 2004;279:36167–36170. doi: 10.1074/jbc.R400017200. [DOI] [PubMed] [Google Scholar]
  39. Toth J, Racz A, Kaminski PM, Wolin MS, Bagi Z, Koller A. Asymmetrical dimethylarginine inhibits shear stress-induced nitric oxide release and dilation and elicits superoxide-mediated increase in arteriolar tone. Hypertension. 2007;49:563–568. doi: 10.1161/01.HYP.0000256764.86208.3d. [DOI] [PubMed] [Google Scholar]
  40. Weibel ER, Hoppeler H. Exercise-induced maximal metabolic rate scales with muscle aerobic capacity. J Exp Biol. 2005;208:1635–1644. doi: 10.1242/jeb.01548. [DOI] [PubMed] [Google Scholar]
  41. Wolin MS, Burke-Wolin TM, Mohazzab HK. Roles for NAD(P)H oxidases and reactive oxygen species in vascular oxygen sensing mechanisms. Respir Physiol. 1999;115:229–238. doi: 10.1016/s0034-5687(99)00023-7. [DOI] [PubMed] [Google Scholar]
  42. Zhilyaev SY, Moskvin AN, Platonova TF, Gutsaeva DR, Churilina IV, Demchenko IT. Hyperoxic vasoconstriction in the brain is mediated by inactivation of nitric oxide by superoxide anions. Neurosci Behav Physiol. 2003;33:783–787. doi: 10.1023/a:1025145331149. [DOI] [PubMed] [Google Scholar]

Articles from The Journal of Physiology are provided here courtesy of The Physiological Society

RESOURCES