Cardiovascular morbidity is the leading cause of mortality worldwide (Mc Namara et al., 2019). Understanding the molecular mechanisms underlying cardiovascular morbidity is, therefore, of utmost importance to provide new therapeutic insight and preventive strategies to reduce mortality and improve the quality of life for patients. Many great discoveries have been made through in vivo assessment of cardiovascular function. Animal models, particularly mice, are often used for this purpose as they are easily accessible and compatible with research on molecular mechanisms underlying cardiovascular physiology and pathology (Janssen et al., 2002; Zaragoza et al., 2011; Camacho et al., 2016). However, many of the suggested molecular mechanisms important for cardiovascular function are based on studies that are limited to isolated organ functions or basic phenotyping reports on tissue perfusion and blood pressure changes. Thus, a comprehensive cardiovascular functional phenotype is rarely described. Existing models have provided significant improvements to the therapeutic armamentarium against cardiovascular diseases, however, it remains the leading cause of death and disability worldwide (Pandya et al., 2013). Hence, new research modalities and an improvement of existing methods are essential to further progress our understanding of mechanisms underlying cardiovascular morbidity (Wier, 2014; Figtree et al., 2021).

Studying the cardiovascular system is a complex task. The cardiovascular system is a closed system consisting of the heart as a pump and blood vessels in which blood flows through. Thus, the efficiency of the circulation is determined by cardiac function, i.e., contractility and heart rate, in conjunction with vascular diameter changes determining total peripheral resistance (Mayet and Hughes, 2003). This interplay between cardiac and vascular functions is further complicated by tight regulation from numerous humoral factors and the autonomic nervous system under normal physiological conditions (Holmes et al., 2004; Gordan et al., 2015; Song et al., 2015; Paz Ocaranza et al., 2020). Evidently, there is a vast number of regulators of the cardiovascular system complicating the mechanistic assessment of the molecular background underlying cardiovascular function under normal physiological conditions. Apart from biological variation including aging and fitness level, deviation from normal physiological conditions due to pathology, e.g., changed sympathetic activity in hypertension, may further contribute to the complexity of the cardiovascular system (Kougias et al., 2010). All these factors must be considered when extrapolating ex vivo knowledge onto whole body homeostasis.

Most conventional ways to investigate the cardiovascular system lack the integrated assessment. For instance, cardiac function is extensively assessed in vivo through various imaging modalities, e.g., echocardiography and magnetic resonance imaging (Ram et al., 2011; Botnar and Makowski, 2012; Li et al., 2020). Albeit this provides great knowledge on cardiac function in vivo, the results from these studies are often seen in isolation and not in conjunction to any physiological changes in peripheral vascular resistance and blood pressure. Vascular function is routinely studied ex vivo, for instance, through different forms of myography, but this is a highly simplified approach as it does not permit simultaneous assessment of changes in blood flow and transmural pressure that take place in vivo (Mulvany and Aalkjaer, 1990). Perfusion of the hind limb or another vascular bed provides a good insight into changes in the peripheral resistance, but this procedure can only be done under conditions of either constant flow or constant pressure (Bomzon and Naidu, 1985). Many studies conclude on vascular function based on blood pressure measurements (Zhao et al., 2011) or tissue perfusion assessment with imaging techniques (Moroz and Reinsberg, 2018), but the conclusions are limited due to lacking information on cardiac output or volume blood flow and driving perfusion pressure, respectively, which are key components for tissue perfusion. Hence, the conventional models are unable to properly address current pressing issues in cardiovascular physiology. One of the challenges is elucidating the mechanisms of molecular signaling or a pathological state with multifarious cardiovascular effects (Tisdale and Gheorghiade, 1992; Holmes et al., 2004; Song et al., 2015; Nielsen et al., 2019; Paz Ocaranza et al., 2020). Moreover, the concurrent assessment of load-dependent and independent parameters is valuable in all disease states and after drug administration. This can be achieved in an integrated model that makes the separation of intrinsic cardiac parameters and vascular responses feasible.

Consequently, the field of cardiovascular physiology demands a comprehensive model for the integrated assessment of the cardiac and vascular functions and their responses to pharmacological intervention. Considering this, we propose an experimental protocol for the integrated assessment of hemodynamic parameters of anesthetized mice before and after pharmacological intervention. By fully elucidating the cardiovascular mechanisms, including load-dependent and independent parameters, our method may contribute to providing valuable new therapeutic targets. In this study, we have challenged mice with a selective α₁ adrenoreceptor agonist, phenylephrine, which has been described to affect both cardiac and vascular functions in vivo (Crosley et al., 1951; Malhotra et al., 1998). We aimed to validate our experimental protocol by distinguishing between phenylephrine-induced vascular and subsequent cardiac changes. We hypothesize that artificial pacing of the heart controls the cardiac function and, hence, can be helpful to differentiate between vascular and cardiac responses. The hemodynamic assessment includes blood pressure, cardiac output, stroke volume, total peripheral resistance, and electrophysiological parameters derived from the electrocardiogram (ECG), and their changes to increasing pacing frequencies.

Instrumentation of blood pressure and flow probes was successful in all six mice, however, ECG data were only available for n = 5. The protocol was completed in all mice. Baseline parameters in the two groups of mice were comparable (Table 1).

This study reports on a mouse model for the assessment of cardiovascular parameters upon in vivo experimental interventions in a comprehensive and integrated manner. This model allows for distinction between interventions that have primary cardiac or vascular actions. We used α₁ adrenoceptor agonist stimulation to modulate vascular function and observed the subsequent effects on overall cardiovascular function. By assessing the response to increasing pacing frequencies compared to periods of rest both with and without phenylephrine, we were able to differentiate between cardiac and vascular effects. We reported complex changes in blood pressure, heart rate, cardiac output, and total peripheral resistance following the interventions.

We chose a selective α₁-adrenergic receptor agonist, phenylephrine, as it is known to increase blood pressure via elevation of systemic vascular resistance without any direct effect on myocardial contractility (Thiele et al., 2011). Phenylephrine is often used clinically as a vasopressor to treat arterial hypotension induced by general anesthesia (Kalmar et al., 2018). Moreover, phenylephrine is broadly used in ex vivo organ bath assessments of vascular functions (Gutierrez et al., 2019), but also in vivo in rodent experimental models (Boguslavskyi et al., 2021; Ralph et al., 2021) where it is used to provoke blood pressure changes. The α₁-adrenergic mechanism underlying the vasopressor effect of phenylephrine is well-described (Dessy et al., 1998; Jumrussirikul et al., 1998; Brozovich et al., 2016; Boguslavskyi et al., 2021). Accordingly, we suggest that the significant phenylephrine-induced elevation of blood pressure in this study is a result of vascular constriction with consequent increase in total peripheral resistance. Furthermore, these elevations in blood pressure and total peripheral resistance were accompanied by a decrease in heart rate. We attributed this heart rate change to be a response to increased activation of baroreceptors, i.e., baroreflex. Upon blood pressure induced activation, the baroreceptors, mediated through the activity of the autonomic nervous system, exercise changes to both cardiac and vascular functions, thus, counteracting any acute pressure changes (Andresen and Kunze, 1994; Jumrussirikul et al., 1998; Boguslavskyi et al., 2021). In our experiments, the baroreceptors failed to return the blood pressure to resting values following phenylephrine injection. This is most likely because phenylephrine-mediated α₁-adrenergic activation of vascular smooth muscle cells bypassed the baroreceptor-mediated inhibition of sympathetic outflow. As a result, phenylephrine maintained vasoconstriction increasing total peripheral resistance and blood pressure. These phenylephrine-induced changes in blood pressure, total peripheral resistance, and heart rate are also described in humans (Crosley et al., 1951; Malhotra et al., 1998).

Phenylephrine administration increased stroke volume. We suggest that this may be result of both increased preload because of phenylephrine-induced venous constriction (Jumrussirikul et al., 1998; Kalmar et al., 2018; Boguslavskyi et al., 2021) and a baroreceptor mediated reduction in heart rate with subsequent increased diastolic filling time. However, this stroke volume elevation is limited by simultaneous elevation of afterload due to increased total peripheral resistance as described previously (Yang et al., 2013; Kalmar et al., 2018). As a product of stroke volume and heart rate, cardiac output did not change significantly. Studies in humans reported both decreased and increased cardiac output following phenylephrine administration because of either increased afterload or reduced heart rate and an increased preload, respectively (Mon et al., 2017; Kalmar et al., 2018). A previous study in pigs suggested this bi-directional effect of phenylephrine to be related to changes to preload and, thus, preload level (Cannesson et al., 2012). Hence, in situations with significantly reduced end-diastolic ventricular filling, e.g., hemorrhage and hypotension, phenylephrine increases preload with consequent elevation in stroke volume and cardiac output (Cannesson et al., 2012). This is supported by observations of increased cardiac output following phenylephrine administration to hypotensive patients (Kalmar et al., 2018). This effect of phenylephrine is predominantly mediated by venous constriction and elevation of venous return leading to increased preload and stroke volume. The lack of a significant increase of cardiac output in this study may suggest that these mice were under relatively mild hypotensive conditions, or that the relation between preload-dependence and cardiac output in mice is different in comparison with humans and pigs (Georgakopoulos and Kass, 2001; Cannesson et al., 2012; Mon et al., 2017; Kalmar et al., 2018).

Mice have high resting heart rates ranging from 500 to 700 beats per minute depending on the time of day (Kaese and Verheule, 2012) corresponding to a very short cardiac cycle length of 80–110 ms. Thus, minimal changes in heart rate can have significant consequences for diastolic duration, diastolic filling time, and stroke volume. The importance of increased ventricular filling time for stroke volume in mice is debated (Georgakopoulos and Kass, 2001). In our study, stroke volume and cardiac output decreased significantly with increasing pacing frequencies suggesting that the ventricular filling time is indeed important. Additionally, the observed difference in stroke volume between the phenylephrine and vehicle groups was abolished during electrical pacing. This implies the importance of reduced heart rate and increase in diastolic filling time for the phenylephrine-induced increments in stroke volume in mice.

A load-dependent biphasic force-frequency relation has previously been described for the murine heart with a positive relation at 400–600 beats per minute (bpm) and a negative force-frequency relation above 600 bpm (Georgakopoulos and Kass, 2001). Accordingly, we saw a negative force-frequency relation during the first pacing session when we paced the hearts with increasing pacing frequencies starting from 10 Hz that corresponds to 600 bpm (Supplementary Figure 1). However, the force-frequency relations were significantly changed for the vehicle and phenylephrine groups during the second pacing session (Supplementary Figure 1). This change may be due to volume loss leading to compensatory endogenous catecholamine release from the sympathetic nerves and adrenal medulla with consequent increase in venous return (Funk et al., 2013). Thus, alterations in preload may have changed the force-frequency relation in our model. Normalizing the murine biphasic force-frequency relation to preload has been suggested to flatten the force-frequency relation curve at high frequencies (Georgakopoulos and Kass, 2001). This may in part explain the decrease in blood pressure seen in both vehicle and phenylephrine groups during the first pacing session and the diminished effect of increasing pacing frequency during the second pacing session. The component of venous return may have been greater following phenylephrine administration compared to vehicle when taking into account phenylephrine-induced venous constriction (Cannesson et al., 2012). Further measurements of pressure-volume relations in the caval vein or right atrium would answer this question.

The versatility of this model should be stated. For instance, this in vivo model can further be elaborated to include myocardial ischemia and reperfusion with temporary occlusion of a coronary artery and, thus, the systemic effects of acute myocardial infarction can be investigated (Michael et al., 1995). Additionally, the protocol may also be applied to other animal models, e.g., the rat model. However, considerations must be taken in regards of different electrophysiological properties of the heart beckoning other settings for the electrical pacing. This remains to be validated in the specific animal models.

Our protocol can distinguish between vascular and cardiac changes and possibly answer important questions in cardiovascular research. This can be particularly pertinent to disease states with multimorbid cardiovascular pathology. For example, cardiovascular morbidity in diabetes and hypertension can comprise pathological changes to both cardiac and vascular functions and lead to heart failure with preserved ejection fraction (Borlaug, 2020). Differentiating between the intrinsic cardiac properties and vascular function may shed light on the mechanism underlying this type of heart failure potentially improving treatment strategies. Currently, this can be examined by a combination of in vivo and ex vivo models, while our protocol offers a purely in vivo setting (Pedersen et al., 2018). Another important issue in cardiovascular research relates to molecules, e.g., ketone bodies and cardiac glycosides, that exert complex responses in the body (Tisdale and Gheorghiade, 1992; Nielsen et al., 2019). Ketone bodies have been described to increase cardiac output (Nielsen et al., 2019). However, due to possible multifarious effects on the cardiovascular system, it is unclear whether this increase in cardiac output is (i) due to increase in contractility or a decrease in heart rate, i.e., a direct cardiac effect, (ii) due to reduced total peripheral resistance, i.e., a direct vascular effect, or (iii) due to a combination of both vascular and cardiac effects. Thus, our model may be a helpful approach to illuminate such intricate research questions.

This experimental protocol provides a possibility to assess valuable comprehensive information on the cardiovascular system, but some limitations must be highlighted and considered for future elaboration of the model. The limited translational value of mouse studies must be considered (Perlman, 2016), yet, mouse experimental models are indispensable in mechanistic studies with genetic expression manipulations (Chinwalla et al., 2002) as well as for initial pharmacological trials and other high throughput analyses (Perlman, 2016). Presently, multiple animal studies must be performed to gain comprehensive knowledge on the effects of a pharmacological treatment. In this regard, our protocol is helpful in reducing the number of animal experiments needed for the comprehensive assessment of in vivo pharmacological interventions in the cardiovascular system.

The established protocol is technically challenging as it is an invasive procedure and requires the introduction of several probes. In this regard, smaller animals may yield a lower surgical success rate due to lower surgical accessibility and increased vulnerability compared to bigger animals. This demands higher surgical skill level and precision. In our experiments, we used 9 months old mice which had suitable size for the surgery required.

Several other aspects related to the surgical procedure must also be mentioned. Opening the thorax alters pressure dynamics in the thoracic cave, e.g., pressure alterations related to natural respiration are eliminated, and this may affect venous return and cardiac function. Additionally, anesthesia and open-thorax interventions have previously been reported to lead to a significant drop in blood pressure (Hoit et al., 1997; Janssen et al., 2004). To reduce the significance of mechanical ventilation between the vehicle and phenylephrine groups, the ventilation parameters were maintained throughout the entire experiment and were comparable between the two groups. In addition, parameters such as positive end-expiratory pressure, airway pressure, and flow were monitored and kept constant in all experiments. Partially, these factors are probably also the reason for the slight hypotensive condition during the experiment. The hypotensive conditions can be corrected by vasopressor administration, e.g., phenylephrine. However, considerations must be taken for the possible interference of several cardiovascular parameters including a direct effect on peripheral vascular resistance with indirect effects on cardiac function as has been shown in this study. Lastly, pancuronium bromide was used in this open-thorax protocol as a muscle relaxant to eliminate intrinsic respiratory reflexes. Pancuronium has been reported to increase heart rate and contractility in rat atrial preparations (Gursoy et al., 2011). Similarly, heart rate but also blood pressure has been described to increase following administration of pancuronium in anesthetized patients (Pauca and Skovsted, 1981). However, we did not see any significant cardiovascular changes in response to intraperitoneal injection of pancuronium used in our study.

Overdriving of heart rate with electrical pacing is not a normal physiological condition. Electrical pacing of the heart with increasing frequency is commonly used in arrhythmia research, e.g., programmed electrical stimulation protocols. However, the pacing settings in our study did not modify the electrical properties of the heart nor introduce arrhythmias evident from ECGs. Electrical pacing successfully and consistently controlled the heart rhythm in the frequency range of 10–11.3 Hz corresponding to a physiological heart rate of 600–678 BPM. Pacing with frequencies below 10 Hz or above 11.3 Hz did not always translate into a 1:1 capture of the heart rhythm. Moreover, recovery of hemodynamic parameters after electrical pacing indicates that cardiovascular functions were not damaged compared to their normal physiological state. Lastly, our results suggest that the electrical pacing had no direct effect on vascular function.

We measured blood flow in the ascending aorta as surrogate for cardiac output. This did not include the blood flow through the coronary arteries and, thus, underestimates the true value of cardiac output. Moreover, change in venous return is an important parameter in our mechanistic assessment, yet, this was not measured directly as this will further complicate the setup. These improvements, e.g., venous blood flow or an atrial pressure-volume assessment, would be an advantage for future technical development. Lastly, despite the importance of neurohumoral regulation for the cardiovascular system, this was not assessed in this study, but this can be considered in future research.

In conclusion, this in vivo protocol provides a method of controlling heart rate in cardiovascular studies in the murine model with pharmacological intervention. Phenylephrine administration increased total peripheral resistance, blood pressure, and stroke volume. However, during atrial pacing there were no differences in stroke volume and cardiac output between the phenylephrine and vehicle groups. Thus, our model suggests that the decrease in heart rate with consequent increase in ventricular filling may be a significant contributor to the phenylephrine-induced increase in stroke volume rather than increase in venous return. This model underlines the primary vascular effects of phenylephrine with secondary cardiac changes. Although technically challenging, this protocol enables the assessment of a broad range of parameters for an integrated understanding of the cardiovascular system under normal and pathological conditions, and it may be helpful in answering important questions in cardiovascular physiology.

The animal study was reviewed and approved by the raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

The experimental protocol was approved by the Animal Experiments Inspectorate of the Danish Ministry of Environment and Food and reported in accordance with the ARRIVE (Animal Research: Reporting in vivo Experiments) guidelines.

RR and VM performed the experiments and made the draft of the manuscript. RR, TP, and VM performed the data analysis. RR, TP, MT, and VM took part in the conception and designed the study. All authors provided critical feedback and approved the final version of the manuscript.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.