INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

CARDIOVASCULAR AND RESPIRATORY drive during exercise originates from two primary mechanisms: central command and reflexes originating in exercising muscle (22, 35). During exercise, muscle reflexes are activated by mechanical and metabolic stimuli that evoke increases in blood pressure, heart rate, ventilation, and region-specific sympathetic nerve activity (13). Muscle contraction elicited by ventral root simulation in anesthetized cats results in a contraction-intensity-dependent increase in cardiovascular and respiratory activity (20). Similar responses are also generated during direct somatic nerve stimulation in the absence of muscle contraction (27). In the case of muscle contraction, cutting the dorsal spinal nerve roots abolishes the responses, demonstrating that the increased activity is generated during muscle contraction and not from direct nerve stimulation (20, 39). Blockade of thinly myelinated (type III) and unmyelinated (type IV) muscle afferent fibers also eliminates the cardiovascular responses to muscle contraction, suggesting that these fiber types are primarily responsible for the evoked responses (20, 21). Static or intermittent static muscle contractions evoked by spinal ventral root stimulation and dynamic muscle contractions elicited by stimulating locomotor regions in the midbrain are able to evoke significant increases in type III and IV muscle afferent discharge from the triceps surae muscles (1, 14). Type III and IV primary muscle afferent fibers carry transduced neural output from the muscle mechano- and metabosensitive receptors to upper lamina dorsal horn neurons of the spinal column (38). Secondary neurons in the dorsal horn project to more rostral regions of the central nervous system involved in regulating cardiovascular and respiratory output (35). Therefore, the central nervous system plays a critical role in mediating the expression of cardiovascular and respiratory drive from contracting muscles.

Several animal models have been developed to study the cardiorespiratory responses to muscle contraction, including cats (5, 20), dogs (8), rats (33, 34), rabbits (31, 37), chickens (29), and humans (6, 9, 12). Wide ranges of cardiovascular responses have been observed in various animal species. Anesthetized cats, dogs, and chickens respond to muscle contraction with an increase in blood pressure and heart rate (5, 8, 20, 29). Humans also increase blood pressure and heart rate during electrically evoked muscle contraction (6, 9, 12). In contrast, anesthetized and decerebrate rabbits respond to muscle contraction with a biphasic response: blood pressure initially decreases and then increases above resting values (31, 37). Various reports show that anesthetized rats respond to muscle contraction with inconsistent changes or no change in arterial pressure that may be dependent on variables such as anesthetic (33, 34). The discrepancy between responses in some animal species compared with humans has limited their role in studying neural regulation of the cardiovascular system during exercise.

Recent studies have demonstrated the ability to chronically record cardiorespiratory activity in conscious mice (19). For example, Desai and colleagues (7) reported increases in blood pressure, heart rate, and ventilation in conscious mice to graded levels of treadmill exercise. These authors found that ventilation and heart rate similar to those in humans were linearly related to workload. Therefore, the mouse appears to be a good model for human cardiovascular and ventilatory responses to dynamic exercise. The utility of this animal species for studying acute, neural regulation of cardiorespiratory function to various autonomic stimuli in the anesthetized state has also begun to emerge (3, 23). These studies provide critical baseline physiological data about the anesthetized mouse as an experimental model and demonstrate the feasibility of utilizing this species in whole animal cardiorespiratory experimentation.

An attractive reason for utilizing the mouse as an experimental animal lies apart from cost effectiveness and availability. Mice have become the predominant species for creating transgenic, knockout, and other genetically manipulated strains. As a result, it is becoming increasingly important to systemically and functionally characterize newly produced mouse strains to more fully understand the physiological impact of genomic manipulation. The molecular techniques used to create genetically altered strains of mice could also provide powerful tools to further the understanding of the influence of single or multiple genes on cardiorespiratory responses to exercise. In a recent example of this approach, knockout mice with a null expression of the myoglobin gene did not show any deficit to exercise or respiratory impact during hypoxia, indicating that myoglobin does not play a critical role in mediating these effects (10).

The same techniques could be used to provide insight into the genomic impact on neural regulation of the cardiorespiratory responses to muscle contraction. However, it is unknown how anesthetized mice respond to muscular contraction. Therefore, the purpose of this study was to measure the cardiovascular responses to muscular contraction, as well as other autonomic stimuli, in mice and evaluate the usefulness of an anesthetized mouse preparation. Our results indicate that muscle contraction in mice elicits significant intensity-dependent increases in cardiovascular functioning. In the same animals, direct somatic nerve stimulation, baroreceptor loading, or systemic hypoxia also elicited appropriate autonomic responses. From these findings, we believe that the anesthetized mouse model is a useful tool for studying the influence of individual genes on reflex regulation of the cardiovascular system to exercise-related stimuli.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Animals. All procedures outlined in this study conform to the National Institutes of Health Guide for the Care and Use of Laboratory Animals. The University of Illinois Animal Care Committee approved all procedures described in these studies. Experiments were performed on adult, male mice (Balb-c, Harlan; n = 12) weighing 24.1 ± 0.4 g. All animals were kept on a 12:12-h light-dark schedule, housed in standard rodent cages, and allowed food and water ad libitum.

Surgical preparation. Mice were initially anesthetized with an intraperitoneal injection of -chloralose (65 mg/kg) and urethan (800 mg/kg). A jugular cannula (PE-10 tubing) was inserted for injection of vasoactive drugs and supplemental anesthetic. Supplemental anesthetic was administered, as needed, when a withdrawal reflex to foot pinch was present. A carotid artery was cannulated with polyethylene tubing (10 gauge) to measure pulsatile arterial pressure (Statham), and heart rate was derived from the pressure signal (Gould Biotach). The small-gauge tubing was connected to a section of larger-gauge (PE-50) tubing before being connected to the transducer, and the shortest possible cannula length was used to help prevent pulse pressure dampening. A tracheotomy was performed to allow the animals to spontaneously breathe room air supplemented with 100% O₂ blown over the end of the tracheotomy tube. Body temperature was measured with a rectal probe (Yellow Springs Instruments) and maintained between 37 and 38°C with a water-perfused heating pad and radiant heat lamp.

Experimental protocols. Animals were allowed to recover from surgery for 30 min before experiments were performed. Basal cardiovascular data were collected for 2 min before muscle contractions were evoked. The cardiovascular responses to muscle contraction were then recorded for 30 s, as were 2 min of recovery data. Static muscle contractions were induced by stimulating the sciatic nerve for 30 s at one to two times motor threshold (MT) with square-wave pulses (0.1-ms duration, 40-Hz frequency). MT was defined as the minimum current required to evoke a muscle twitch. Mice were allowed to recover for 15 min between trials. Muscle contractions were evoked a maximum of three times in each animal. Two contractions were performed utilizing voltages twice the MT to test for repeatability of cardiovascular responses. A third contraction was evoked at MT to determine the relationship between muscle tension and cardiovascular responses.

Data analysis. A paired Student's t-test was utilized to test for changes from baseline in mean arterial pressure and heart rate. Test-retest reliability of cardiovascular responses to muscle contraction was also determined with a Pearson's product-moment correlation coefficient and paired t-test. We examined the best-fit relationship (linear regression) from a scatterplot of all the data points and derived a force-response relationship across animals for the arterial pressure responses to muscular contraction. A Student's paired t-test was utilized to test for differences in arterial pressure responses to two different levels of muscle contraction (MT vs. 2 times MT). Linear regression techniques were also used to determine the best-fit relationships between arterial pressure and heart rate changes during somatic nerve stimulation and baroreflex testing (intravenous phenylephrine injections). Values are means ± SE. All statistical tests were deemed to be significant at P < 0.05.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Muscle contraction. Sciatic nerve stimulation evoked muscle contraction in mice (n = 7) that resulted in significant increases in cardiovascular activity. Figure 1A shows traces of the cardiovascular responses in one mouse to muscle contraction. Muscle contraction quickly elicited increases in arterial pressure and heart rate (<5 s) that peaked at approximately the same time as muscle force and returned to precontraction values before the contraction was completed. Average increases in mean arterial pressure and heart rate for all contractions were 13.8 ± 1.9 mmHg (range 6-25 mmHg) and 33.5 ± 11.9 beats/min (range 15-110 beats/min), respectively (both P < 0.05). Average resting and peak arterial pressures and heart rates during muscle contraction are depicted in Fig. 1B. In all cases, cardiovascular responses were abolished after distal sciatic nerve crush or cut, indicating that the responses were generated by muscular contraction. Test-retest comparisons (n = 5) yielded a significant correlation (r = 0.77, P < 0.05) for the mean arterial pressure responses between the first (10.4 ± 2.6 mmHg) and second (9.9 ± 2.9 mmHg) muscle contractions. Multiple muscle contractions also generated consistent tension values (102.1 ± 22.3 and 100.1 ± 26.9 g for first and second trials, respectively, r = 0.83, P < 0.05). Paired Student's t-tests did not detect any difference between the first and second contraction forces or blood pressure responses. These data indicate that the cardiovascular responses during muscle contraction in mice are repeatable physiological events. Linear regression analysis of peak muscle tensions vs. peak arterial pressure responses for all animals tested (n = 7) yielded an R² value of 0.2269. However, a significant contraction intensity dependence of arterial pressure response (13.8 ± 2.2 vs. 6.3 ± 1.8 mmHg) was noted between two different contraction intensities (122 ± 31 g for 2 times MT vs. 77 ± 29 g for MT) from all mice that performed muscle contractions. In ~20% of the total mice, we were not able to generate cardiovascular responses to muscle contraction. These animals were subsequently excluded from the study. However, all these animals had very low resting arterial pressures (<30 mmHg), and the lack of reflex cardiovascular response most likely reflected the preparatory state of the animal.

View larger version (22K): [in this window] [in a new window]   Fig. 1.   A: cardiovascular responses to muscle contraction. Muscle contraction was evoked at 2 times motor threshold (MT) (0.1-ms square-wave pulse, 40 Hz) and significantly elevated arterial pressure and heart rate. Crushing the nerve distal to the stimulating electrode eliminated the cardiovascular responses to nerve stimulation. This shows that the responses were not evoked by direct afferent nerve stimulation but rather by muscle contraction. B and C: average arterial pressure and heart rate responses, respectively. Values are means ± SE. *P < 0.05 (paired t-test).

Somatic stimulation. Stimulation of the crushed (or cut) central end of the sciatic (n = 7), peroneal (n = 6), tibial (n = 5), or sural (n = 4) nerves (0.5-40.0 mV) resulted in significant increases in mean arterial pressure and heart rate. Since muscular contraction was not observed during these tests, increases in blood pressure were driven through direct electrical activation of somatic afferent fibers. Maximal mean arterial pressure responses to tibial, peroneal, sural, and sciatic nerve stimulation were 23.1 ± 3.5, 20.9 ± 4.0, 21.7 ± 2.6, and 25.8 ± 3.0 mmHg, respectively (all P < 0.05). Maximal heart rate responses to tibial, peroneal, sural, and sciatic nerve stimulation were 29.3 ± 8.5, 25.6 ± 5.0, 48.0 ± 7.1, and 35.9 ± 10.6 beats/min, respectively (all P < 0.05). Cardiovascular responses from all somatic nerves stimulated were intensity dependent. The Pearson's product-moment correlation coefficients relating the mean arterial pressure response to nerve stimulation intensity (percentage of minimal response) for sciatic, tibial, peroneal, and sural nerves were 0.53, 0.75, 0.76, and 0.81, respectively (all P < 0.05).

Arterial baro- and chemoreflexes. Intravenous injections of phenylephrine (n = 8) induced an average (all trials) increase in arterial pressure of 68 ± 13.5 mmHg and a strong reflexive decrease in heart rate of 158.2 ± 18.2 beats/min (Fig. 2A). Heart rate responses to phenylephrine-induced increases in arterial pressure were also graded (Fig. 2B). Increasing doses of phenylephrine and a subsequent rise in pressure produced greater drops in heart rate. Hypoxia (n = 9) also induced increases in cardiovascular activity (Fig. 3). Breathing 10% O₂-90% N₂ for 20-30 s resulted in an increase of mean arterial pressure by 11.7 ± 2.6, and an increase of heart rate by 42.7 ± 13.9 beats/min.

View larger version (22K): [in this window] [in a new window]   Fig. 2.   Baroreflex testing was evaluated by increasing arterial pressure with intravenous injections of phenylephrine (10-50 µg/kg). A: raw data trace demonstrating a reflex bradycardia to a phenylephrine-induced increase in arterial pressure. B: graded decreases in heart rate were observed across animals with increasing levels of mean arterial pressure (MAP). Linear regression techniques were used to describe the line of best fit.

View larger version (24K): [in this window] [in a new window]   Fig. 3.   Cardiovascular responses to chemoreflex testing in anesthetized mice. A: raw trace showing arterial pressure and heart rate responses to 30 s of hypoxia (10% O2-90% N2). B: switching to 10% O2-90% N2 induced significant elevations of mean arterial pressure and heart rate. Values are means ± SE. *P < 0.05

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The results from the present study indicate that anesthetized mice respond to several autonomic stimuli, including exercise-related stimuli, such as muscle contraction, with an increase in cardiovascular activity. Arterial chemo- and baroreflexes are also able to homeostatically regulate the cardiovascular system in this preparation. Moreover, we have demonstrated the feasibility of performing muscle pressor reflex studies with anesthetized mice and consider this species to be a useful model for studying the mechanisms responsible for neural regulation of the cardiovascular system during exercise.

Recent studies have characterized the cardiovascular and respiratory responses to exercise in mice (7). The authors found that mice and humans demonstrate similar responses to dynamic exercise on a treadmill and that the responses were graded according to exercise intensity. During voluntary exercise, two primary neural mechanisms are activated to increase cardiorespiratory activity: central command and muscle reflexes. Also, several other feedback mechanisms, such as arterial chemoreflexes, baroreflexes, and vestibuloautonomic reflexes, can integrate with central command and muscle reflex drive to provide a more coordinated cardiorespiratory response to exercise (16, 25). We have demonstrated, for the first time, that muscle reflexes can drive cardiovascular function in mice and that arterial chemo- and baroreflexes are also present in the same animals. As a result, mice appear to be a useful model for studying central and peripheral autonomic integration.

Other anesthetized rodent models of cardiovascular regulation during muscle contraction have yielded inconsistent results. Different authors have described decreases (33) or no change (34) in arterial pressure in response to muscle contraction in the anesthetized rat. Toney and Mifflin (33) reported a decrease in arterial pressure with intermittent static contractions but not sustained static contractions. Similarly, Vissing et al. (34) demonstrated no change in arterial pressure with sustained, tetanic contractions. In the latter study, an increase in sympathetic nerve activity was noted during muscle contraction in anesthetized rats, while Toney and Mifflin noted a blunted depressor response after sectioning of adrenal sympathetic nerves or removal of the adrenal glands. These findings differ from that observed in the anesthetized cat and dog. Moreover, electrically evoked static muscle contractions in humans also increase blood pressure (6, 9, 12). Thus the repeatable increases in blood pressure in response to muscle contraction in the anesthetized mouse are consistent with other mammal species.

Muscle contraction in the anesthetized mouse evoked increases in blood pressure that were logarithmically related to peak muscle tension. Coote and colleagues (5) also demonstrated a logarithmic relationship between peak muscle tension and arterial pressure responses in anesthetized cats. It is unclear from this study whether increased muscle tension or increased activation of muscle mass was responsible for the graded increase in blood pressure, but these responses are congruent with intensity-dependent increases in cardiovascular activity in humans performing isometric muscle contractions (22) and mice performing treadmill exercise (7). The linear regression analysis relating peak muscle tension developed to peak arterial pressure response for all contractions yielded a relationship that was relatively weak (R² = 0.2269). However, a paired t-test analysis of the blood pressure responses to two different contraction intensities within all mice tested demonstrated a significant arterial pressure dependence on peak muscle tension generated. This discrepancy is likely explained by the fact that the regression analysis contained inter- and intra-animal errors, while the paired t-test only accounted for a within-animal error. Conversely, direct afferent stimulation of all nerves tested showed a strong linear relationship (regression) to the evoked cardiovascular responses.

High-frequency stimulation of the tibial nerve in anesthetized rats evokes decreases in arterial pressure, while stimulating the sural nerve evokes increases in blood pressure (26, 30). This is different from the anesthetized cat, where stimulation of the tibial nerve evokes increases in blood pressure (27, 36). Previously, Butcher and colleagues (3) recorded increases in arterial pressure and heart rate with "noxious" foot or tail pinch in anesthetized mice. Our results concur with their previous findings of increased cardiovascular activity with general somatic stimulation. However, our data extend these findings by further showing that stimulating nerves containing afferent fibers originating in predominantly different tissues (i.e., muscle or skin) induces increases in cardiovascular activity.

It was demonstrated in this study that arterial baro- and chemoreflexes evoke cardiovascular responses in the anesthetized mouse. These findings are in agreement with a previous study with mice (18). These authors recorded cardiovascular and sympathetic nerve activity in urethan-anesthetized mice and demonstrated graded heart rate and renal sympathetic nerve responses to baroreceptor stimulation. The authors also reported significant elevations in renal sympathetic nerve activity with hypoxic stimulation that are consistent with our observations of elevated blood pressure. The mouse preparation that we have described here also provides for multiple autonomic stimuli to be tested in the same animal. This is beneficial, in that it allows the potential to study the integration of several autonomic regulatory mechanisms.

The resting blood pressures of the mice utilized in this study were relatively low compared with previously published studies in anesthetized mice (3, 18). One difference between this study and those previously published was that we utilized a combination of -chloralose and urethan, while only urethan was used previously. However, despite the anesthetic-induced depression of resting arterial pressure, the mice were able to generate robust cardiovascular responses to autonomic stimuli.

Mice are the most prevalent species for creating genetically altered strains of animals. As a result, murine models have become a powerful tool to understand the genomic impact on cardiovascular regulation (28, 32). Several knockout strains of mice have been physiologically characterized in response to various individual autonomic stressors, including systemic hypoxia, arterial baroreceptor challenge, and exercise (3, 10, 15, 17, 24). Potential mouse candidates to help identify genes involved in regulating cardiovascular function during muscle contraction include recently developed ₂-receptor and capsaicin (vanilloid) knockout mice (2, 4, 11). Each of these examples represents central (₂-subtypes) and peripheral (capsaicin receptor) factors that may play a role in mediating the cardiovascular responses to muscle contraction. These strains of mice and the many genetically manipulated mice likely to follow present genotypes that allow for the phenotypic characterization of elements involved in cardiovascular responses to muscle contraction. Certainly, as with any experimental model, there are limitations that one must bear in mind when utilizing genetically manipulated mice, including developmental issues and physiological redundancy. When carefully selected, however, these mouse strains offer a unique opportunity to study the role that single genes may play in mediating the cardiovascular responses to exercise-related stimuli, including muscle contraction. Potentially, genetically altered mouse strains can help identify key factors involved in peripheral and central regulation of cardiorespiratory responses to exercise. Moreover, genetically manipulated animals may be useful in furthering our understanding of the impact of acute and chronic exercise on gene regulation and subsequent alterations in physiological function.