BRIEF RESEARCH REPORT published: 09 January 2020 doi: 10.3389/fneur.2019.01364 Chronic Sleep Fragmentation Mimicking Sleep Apnea Does Not Worsen Left-Ventricular Function in Healthy and Heart Failure Mice Ignacio Cabrera-Aguilera 1,2 , Begoña Benito 3,4,5 , Marta Tajes 4 , Ramon Farré 1,6,7 , David Gozal 8 , Isaac Almendros 1,6,7 and Nuria Farré 4,5,9* 1 Unitat de Biofísica i Bioenginyeria, Facultat de Medicina i Ciències de la Salut, Universitat de Barcelona, Barcelona, Spain, 2 Departament of Human Movement Sciences, Faculty of Health Sciences, School of Kinesiology, Universidad de Talca, Talca, Chile, 3 Department of Cardiology, Hospital del Mar, Barcelona, Spain, 4 Heart Diseases Biomedical Research Group, IMIM (Hospital del Mar Medical Research Institute), Barcelona, Spain, 5 Department of Medicine, Universitat Autònoma de Barcelona, Barcelona, Spain, 6 CIBER de Enfermedades Respiratorias, Madrid, Spain, 7 Institut d’Investigacions Biomèdiques August Pi i Sunyer, Barcelona, Spain, 8 Department of Child Health and Child Health Research Institute, The University of Missouri School of Medicine, Columbia, MO, United States, 9 Heart Failure Unit, Department of Cardiology, Hospital del Mar, Barcelona, Spain Edited by: Aims: Obstructive sleep apnea (OSA) has been associated with heart failure (HF). Victor B. Fenik, Sleep fragmentation (SF), one of the main hallmarks of OSA, induces systemic VA Greater Los Angeles Healthcare System, United States inflammation, oxidative stress and sympathetic activation, hence potentially participating Reviewed by: in OSA-induced cardiovascular consequences. However, whether SF per se is Sina A. Gharib, deleterious to heart function is unknown. The aim of this study was to non-invasively University of Washington, United States evaluate the effect of SF mimicking OSA on heart function in healthy mice and in mice Jimmy Fraigne, with HF. University of Toronto, Canada Methods and Results: Forty C57BL/6J male mice were randomized into 4 groups: *Correspondence: Nuria Farré control sleep (C), sleep fragmentation (SF), isoproterenol-induced heart failure (HF), and NFarreLopez@parcdesalutmar.cat mice subjected to both SF+HF. Echocardiography was performed at baseline and after 30 days to evaluate left ventricular end-diastolic (LVEDD) and end-systolic (LVESD) Specialty section: This article was submitted to diameters, left ventricular ejection fraction (LVEF) and fraction shortening (FS). The effects Sleep Disorders, of SF and HF on these parameters were assessed by two-way ANOVA. Mice with a section of the journal isoproterenol-induced HF had significant increases in LVEDD and LVESD, as well as a Frontiers in Neurology decreases in LVEF and FS (p = 0.013, p = 0.006, p = 0.027, and p = 0.047, respectively). Received: 21 October 2019 Accepted: 10 December 2019 However, no significant effects emerged with SF (p = 0.480, p = 0.542, p = 0.188, and Published: 09 January 2020 p = 0.289, respectively). Citation: Cabrera-Aguilera I, Benito B, Tajes M, Conclusion: Chronic SF mimicking OSA did not induce echocardiographic changes Farré R, Gozal D, Almendros I and in cardiac structure and function in both healthy and HF mice. Thus, the deleterious Farré N (2020) Chronic Sleep cardiac consequences of OSA are likely induced by other perturbations associated Fragmentation Mimicking Sleep Apnea Does Not Worsen Left-Ventricular with this prevalent condition, or result from interactions with underlying comorbidities Function in Healthy and Heart Failure in OSA patients. Mice. Front. Neurol. 10:1364. doi: 10.3389/fneur.2019.01364 Keywords: sleep apnea, sleep breathing disorders, heart failure, ventricular function, sleep fragmentation Frontiers in Neurology | www.frontiersin.org 1 January 2020 | Volume 10 | Article 1364 Cabrera-Aguilera et al. Sleep Fragmentation and Heart Failure INTRODUCTION Echocardiography (Vivid IQ and L8-18i-D Linear Array 5– 15 MHz, General Electric Healthcare, Horten, Norway) was Sleep apnea is the most frequent sleep breathing disorder in performed in each animal at baseline, before surgery, and patients with heart failure (HF), and is strongly associated with at day 30 following a standard protocol (9). The following worse prognosis (1). The two major physiological perturbations echographic indices were subsequently computed: left ventricular experienced by patients with sleep apnea are intermittent end-diastolic (LVEDD) and end-systolic diameter (LVESD), left hypoxia and sleep fragmentation (SF), and are a consequence ventricular ejection fraction (LVEF) and fraction shortening (FS). of the recurrent events of nocturnal airway collapse. These two All echocardiographic measurements and computations were characteristic perturbations have been mechanistically implicated carried out by a single operator (NF), who was blind to the in the mid-term and long-term adverse consequences of OSA. experimental group. However, whereas the effects of recurrent hypoxemia in the Immediately after the final echography, blood was extracted healthy heart or in HF have been extensively studied (2, 3), there to measure circulating interleukin (IL)-6 (ELISA kit Ab100712, is no specific information on the potential cardiac effects induced Abcam), which can be considered a systemic biomarker of SF by chronic SF. Given that increased sympathetic activation is effects (10), and animals were euthanized by exsanguination. common in both HF and SF (4, 5), it is possible that SF negatively All data are presented as mean ± SEM. Comparison of contributes to cardiac function in the healthy heart, and more echocardiographic data between all groups at baseline was particularly in HF. performed using one-way ANOVA. Comparison of IL-6 plasma The aim of this work was to assess whether chronic SF affects concentrations and echocardiographic data between all groups at ventricular cardiac function in a mouse model mimicking sleep day 30 was performed using two-way ANOVA followed by the apnea. To this end, we used non-invasive cardiac echography Student-Newman-Keuls comparison method. For all tests, p < since this technique is immediately applicable to routine 0.05 was considered statically significant. patient assessment. RESULTS METHODS First, as expected from random distribution of mice into the four experimental groups, there were no significant differences The experimental protocol (approved by the institutional Ethical in the baseline values of LVEDD, LVESD, LVEF, and FS (p = Board) was conducted on 40 male mice (C57BL/6J; 10 weeks 0.308, 0.756, 0.817, and 0.771, respectively). Second, effective old; 12 h light/dark cycle; water/food ad libitum) randomly application of SF was confirmed by a 10-fold (p < 0.001; 2-way distributed into 4 groups (n = 10 each). In two groups the ANOVA) increase in circulating IL-6 in the mice subjected to mice were allowed to sleep normally: healthy control (C) and this challenge [82.2 ± 27.0 pg/ml (C-SF) and 62.2 ± 36.3 pg/ml heart failure (HF). In two similar groups (C+SF, HF+SF), SF (HF+SF)], as compared with control sleep mice [9.1 ± 6.4 pg/ml mimicking sleep apnea was imposed. (C) and 6.3 ± 2.8 pg/ml (HF)]. Third, isoprotenerol infusion HF was conventionally induced by continuous infusion of effectively induced HF. As expected from extensive data in the isoproterenol (6). Briefly, mice in HF and HF+SF groups were literature (6), mice in the HF groups had significant increases in anesthetized by isoflurane inhalation and an osmotic mini-pump LVEDD and LVESD as well as significant reductions in LVEF and (Alzet, model 1004) was implanted subcutaneously in the flank. FS (Figure 1). Indeed, 2-way ANOVA of the 4 groups showed The pump was set to deliver 30 mg/kg per day of isoproterenol that HF was significant (p = 0.013, 0.006, p = 0.027, and p = (Sigma Aldrich; in sterile 0.9% NaCl solution) for 28 days. 0.047 for LVEDD, LVESD, LVEF, and FS, respectively). Buprenorphine (0.3 mg/kg, i.p.) was administered 10 min before The main finding however, was that SF did not significantly surgery and after 24 h, and the suture was removed 7 days worsen ventricular function (LVEDD, LVESD, LVEF, and FS), after surgery. Healthy animals (C, C+SF) were subjected to the neither in healthy mice, nor in animals with HF (Figure 1), as same protocol with the only difference that no isoprotenerol was indicated by 2-way ANOVA (p = 0.294, p = 0.657, 0.509, and dissolved into the 0.9% NaCl pump medium. p = 0.625, respectively). Moreover, there were no statistically Two days after surgery, SF was daily applied by means significant interactions between factors HF and SF (p = 0.480, of a previously described and validated device for mice p = 0.542, 0.188, and p = 0.289, respectively). (Lafayette Instruments, Lafayette, IN) based on intermittent tactile stimulation (7, 8). Sleep arousals were induced by a mechanical near silent motor with a horizontal bar sweeping DISCUSSION just above the cage floor from one side to the other side in the standard mouse laboratory cage. This automatic system, with no Although sleep apnea considerably increases the morbidity human intervention, minimized stress to the animal. To apply SF and mortality of cardiovascular diseases (1), the potential mimicking sleep apnea, 2 min intervals between each sweep (i.e., mechanisms are poorly understood. Whereas, the deleterious corresponding to an apnea-hypopnea index of 30 events/h) were role of intermittent hypoxemia has been widely investigated, the applied during the murine sleep period (8 a.m. to 8 p.m.) for 28 cardiac effects of SF, the other main physiological perturbation days (until day 30 from surgery). in sleep apnea, has not been specifically explored to date. For Frontiers in Neurology | www.frontiersin.org 2 January 2020 | Volume 10 | Article 1364 Cabrera-Aguilera et al. Sleep Fragmentation and Heart Failure FIGURE 1 | Sleep fragmentation (SF) mimicking sleep apnea did not modify left ventricular echocardiographically-measured structure and function in control healthy mice (C) and in heart failure (HF) mice. LVEDD, end-diastolic diameter; LVESD, end-systolic diameter; LVEF, left ventricular ejection fraction; FS, fraction shortening. Only HF significantly changed cardiac indices in both normal sleep and in SF (**p < 0.05), with no significant differences induced by SF in healthy controls (C) and in HF mice. the first time, this study provides experimental data showing hypoxia resulted in altered heart function, as measured by that SF mimicking sleep apnea does not worsen cardiac function. echocardiography. Surprisingly, intermittent hypoxia was not Interestingly, this lack of effect of SF was seen both in healthy deleterious, and in fact resulted in improvements in cardiac mice and in animals with HF. function (2, 3). In this setting, we did not find any changes We conducted the study in an isoproterenol-induced induced by SF, even though application of SF of similar duration heart failure model since the involved mechanism [increased has been shown to be deleterious to several organs other than sympathetic activation (11)] could be synergistic with the the heart (7, 8, 10, 13). However, we cannot rule out that a physiological response to SF. Indeed, sleep fragmentation is longer exposure to SF could be required to manifest cardiac considered a main determinant of the sympathetic activation function alterations. The fact that physiological perturbations in sleep apnea, independently of the frequency and severity of characterizing sleep apnea (SF and recurrent hypoxemia) are not oxygen desaturation (4). It should be mentioned, however, that deleterious to cardiac function is intriguing. Besides the need animals in the HF model could have some sleep alterations for expanded research on this issue, the cumulative findings to since it has been reported that direct infusion of isoproterenol date may account for some of the contrasting results reported into the lateral hypothalamic area may result in increased by clinical studies addressing the interaction between sleep apnea arousability (12). However, it is not possible to compare the and heart diseases. Indeed, in obesity hypoventilation syndrome, doses directly injected into the lateral hypothalamic area with the prevalence of chronic heart failure had the strongest negative the ones that correspond into this area from the concentration association with the highest tertile of oxygen desaturation of the intraperitoneally injected isoproterenol in the model of index (14). More intriguingly, treatment of severe central sleep SF. Notwithstanding, to the best of our knowledge no sleep apnea with adaptive servo-ventilation in patients with HF was alterations have been reported from the large number of studies associated with an increase in all-cause and cardiovascular using the isoproterenol-induced heart failure model. Hence, this mortality (15). question deserves further investigation, for instance by EEG- Our experimental study has some limitations. Our based sleep characterization of animals subjected to this drug- measurements do not completely preclude an adverse effect induced heart failure model. However, contrary to what would of SF on the heart, since it is still possible that diastolic function be expected, our study showed that SF mimicking sleep apnea did could be affected. However, the few studies that have reported not affect cardiac function. changes in diastolic function and sleep abnormalities involved It is interesting to compare our results on SF with previous acute sleep deprivation (16), and not SF. The possibility also data on intermittent hypoxia. Studies carried out in C57BL/6J remains that SF could worsen HF of different etiologies, even and HF-transgenic mice reported that 28 days of intermittent if no such effects were apparent in our HF model. Accordingly, Frontiers in Neurology | www.frontiersin.org 3 January 2020 | Volume 10 | Article 1364 Cabrera-Aguilera et al. Sleep Fragmentation and Heart Failure studying how SF interacts with HF in models induced by ETHICS STATEMENT different challenges (e.g., pressure/volume overload or ischemic injury) (17) is warranted. In addition, we evaluated ventricular The animal study was reviewed and approved by Ethical function by echocardiography to assess for clinically relevant Committee for Animal Research of the University of Barcelona. structural and functional changes. However, we cannot rule out whether SF induces minor changes at the molecular and AUTHOR CONTRIBUTIONS cellular level in the heart. It is also worth noting that our study was carried out in male animals exclusively. Taking IC-A carried out the experiments, analyzed the data, and into account that some sex differences have been reported in contributed to draft the manuscript. BB, MT, RF, and DG rodent heart failure models (18), future studies addressing the contributed to analyze the data and revise the manuscript. NF interaction between SF and HF should be also carried out in and IA conceived the work. NF carried out measurements, female mice. drafted the manuscript, and supervised the study. In conclusion, SF did not induce echocardiographic detectable deleterious effects in heart structure and function in a sleep apnea FUNDING model imposed on both healthy and HF mice. These findings suggesting that the potential cardiac consequences of sleep apnea This work was supported in part by the Spanish Ministry of in patients are mainly induced by other perturbations resulting Economy and Competitiveness (SAF2017-85574-R). IC-A was from the obstructive nocturnal events or reflect interactions with supported by CONICYT PFCHA—Chilean Doctorate Fellowship the underlying comorbidities frequently found in patients with 2017; Grant No. 72180089. DG was supported in part by National sleep apnea. Institutes of Health grants HL130984 and HL140548. DATA AVAILABILITY STATEMENT ACKNOWLEDGMENTS All datasets generated for this study are included in the The authors wish to thank Ms. Elisabet Urrea for their excellent article/supplementary material. technical assistance. REFERENCES 12. Schmeichel BE, Berridge CW. Wake-promoting actions of noradrenergic α1– and β-receptors within the lateral hypothalamic area. Eur J Neurosci. (2013) 1. Cowie MR, Gallagher AM. Sleep disordered breathing and heart failure. JACC 37:891–900. doi: 10.1111/ejn.12084 Hear Fail. (2017) 5:715–23. doi: 10.1016/j.jchf.2017.06.016 13. Trzepizur W, Cortese R, Gozal D. Murine models of sleep apnea: 2. Naghshin J, Rodriguez RH, Davis EM, Romano LC, McGaffin KR, O’Donnell functional implications of altered macrophage polarity and epigenetic CP. Chronic intermittent hypoxia exposure improves left ventricular modifications in adipose and vascular tissues. Metabolism. (2018) 84:44– contractility in transgenic mice with heart failure. J Appl Physiol. (2012) 55. doi: 10.1016/j.metabol.2017.11.008 113:791–8. doi: 10.1152/japplphysiol.00185.2012 14. Masa JF, Corral J, Romero A, Caballero C, Terán-Santos J, 3. Naghshin J, McGaffin KR, Witham WG, Mathier MA, Romano Alonso-Álvarez ML, et al. Protective cardiovascular effect of sleep LC, Smith SH, et al. Chronic intermittent hypoxia increases left apnea severity in obesity hypoventilation syndrome. Chest. (2016) ventricular contractility in C57BL/6J mice. J Appl Physiol. (2009) 150:68–79. doi: 10.1016/j.chest.2016.02.647 107:787–93. doi: 10.1152/japplphysiol.91256.2008 15. Cowie MR, Woehrle H, Wegscheider K, Angermann C, d’Ortho 4. Seravalle G, Mancia G, Grassi G. Sympathetic nervous M-P, Erdmann E, et al. Adaptive servo-ventilation for central system, sleep, and hypertension. Curr Hypertens Rep. (2018) sleep apnea in systolic heart failure. N Engl J Med. (2015) 20:74. doi: 10.1007/s11906-018-0874-y 373:1095–105. doi: 10.1056/NEJMoa1506459 5. Słomko J, Zawadka-Kunikowska M, Klawe JJ, Tafil-Klawe M, Newton J, 16. Açar G, Akçakoyun M, Sari I, Bulut M, Alizade E, Özkan B, et al. Zalewski P. Cardiovascular regulation and body temperature: evidence Acute sleep deprivation in healthy adults is associated with a reduction from a nap vs. sleep deprivation randomized controlled trial. Physiol Res. in left atrial early diastolic strain rate. Sleep Breath. (2013) 17:975– (2018). doi: 10.33549/physiolres.933758. [Epub ahead of print]. 83. doi: 10.1007/s11325-012-0786-z 6. Chang SC, Ren S, Rau CD, Wang JJ. Isoproterenol-induced heart failure 17. Riehle C, Bauersachs J. Small animal models of heart failure. Cardiovasc Res. mouse model using osmotic pump implantation. Methods Mol Biol. (2018) (2019) 115:1838–49. doi: 10.1093/cvr/cvz161 1816:207–20. doi: 10.1007/978-1-4939-8597-5_16 18. Regitz-Zagrosek V, Oertelt-Prigione S, Seeland U, Hetzer R. Sex and gender 7. Nair D, Zhang SXL, Ramesh V, Hakim F, Kaushal N, Wang Y, et al. differences in myocardial hypertrophy and heart failure. Circ J. (2010) Sleep fragmentation induces cognitive deficits via nicotinamide adenine 74:1265–73. doi: 10.1253/circj.CJ-10-0196 dinucleotide phosphate oxidase-dependent pathways in mouse. Am J Respir Crit Care Med. (2011) 184:1305–12. doi: 10.1164/rccm.201107-1173OC Conflict of Interest: The authors declare that the research was conducted in the 8. Ramesh V, Kaushal N, Gozal D. Sleep fragmentation differentially modifies absence of any commercial or financial relationships that could be construed as a eeg delta power during slow wave sleep in socially isolated and paired mice. potential conflict of interest. Sleep Sci. (2009) 2:64–75. 9. Gao S, Ho D, Vatner DE, Vatner SF. Echocardiography in mice. Curr Protoc Copyright © 2020 Cabrera-Aguilera, Benito, Tajes, Farré, Gozal, Almendros Mouse Biol. (2011) 1:71–83. doi: 10.1002/9780470942390.mo100130 and Farré. This is an open-access article distributed under the terms of the 10. Carreras A, Zhang SX, Peris E, Qiao Z, Gileles-Hillel A, Li RC, et al. Chronic Creative Commons Attribution License (CC BY). The use, distribution or sleep fragmentation induces endothelial dysfunction and structural vascular reproduction in other forums is permitted, provided the original author(s) changes in mice. Sleep. (2014) 37:1817–24. doi: 10.5665/sleep.4178 and the copyright owner(s) are credited and that the original publication in 11. Kudej RK, Iwase M, Uechi M, Vatner DE, Oka N, Ishikawa Y, et al. Effects of this journal is cited, in accordance with accepted academic practice. No use, chronicβ-adrenergic receptor stimulation in mice. J Mol Cell Cardiol. (1997) distribution or reproduction is permitted which does not comply with these 29:2735–46. doi: 10.1006/jmcc.1997.0508 terms. Frontiers in Neurology | www.frontiersin.org 4 January 2020 | Volume 10 | Article 1364