Irisin evokes bradycardia by activating cardiac-projecting neurons of nucleus ambiguus

Irisin is a newly identified hormone induced in muscle and adipose tissues by physical activity. This protein and its encoding gene have been identified in the brain; in addition, the precursor for irisin, FNDC5, can cross the blood-brain barrier. The fact that irisin is secreted during exercise together with the lower resting heart rate in athletes prompted us to investigate the effect of irisin on cardiac-projecting vagal neurons of nucleus ambiguus, a key regulatory site of heart rate. In vitro experiments in cultured nucleus ambiguus neurons indicate that irisin activates these neurons, inducing an increase in cytosolic Ca2+ concentration and neuronal depolarization. In vivo microinjection of irisin into the nucleus ambiguus promotes bradycardia in conscious rats. Our study is the first to report the effects of irisin on the neurons controlling the cardiac vagal tone and to link a myokine to a cardioprotective role, by modulating central cardiovascular regulation.


Introduction
Irisin is a recently described myokine, secreted in the rodent and human skeletal muscle as a result of exercise (Bostrom et al. 2012). In muscle, exercise promotes the formation of fibronectin type III domain containing 5 (FNDC5, aka PeP, and FRCP2) and its cleavage to irisin by the transcriptional co-activator PPAR-c co-activator-1 a (PGC1a) (Bostrom et al. 2012). Irisin stimulates thermogenesis and browning of white adipose tissue and increases energy expenditure (Bostrom et al. 2012). Irisin was found to be secreted also by the white adipose tissue and considered an adipokine that may contribute to a muscle-adipose tissue regulatory mechanism (Roca-Rivada et al. 2013;Crujeiras et al. 2015).
Besides skeletal muscle and adipose tissue, irisin has been recently identified in several other tissues including the brain (Aydin 2014;Aydin et al. 2014). Irisin was found in discrete regions of the brain and in the cerebrospinal fluid (Dun et al. 2013;Piya et al. 2014). FNDC5 brain expression is increased by exercise; in addition, the precursor for irisin, FNDC5, crosses the blood-brain barrier (Wrann et al. 2013), where it may be converted to irisin.
The circulating levels of irisin and the rodent-tohuman translation of irisin biology have been subject of controversy (Crujeiras et al. 2015;Elsen et al. 2014;Sanchis-Gomar et al. 2014;Albrecht et al. 2015;Kerstholt et al. 2015). Clinical studies have shown increased circulating level of irisin after acute (Huh et al. 2012;Brenmoehl et al. 2014;Loffler et al. 2015) or chronic exercise (Bostrom et al. 2012;Ijiri et al. 2015). Increased serum irisin levels were found in active subjects (Moreno et al. 2015) or healthy centenarians, while young patients with myocardial infarction (Emanuele et al. 2014) or patients with type 2 diabetes (Xiang et al. 2014) had significantly reduced levels of irisin. Other studies found that, plasma irisin levels appear to be negatively correlated with highdensity lipoproteins (HDL) cholesterol levels (Panagiotou et al. 2014), while positively correlated with markers of obesity (Stengel et al. 2013) and metabolic syndrome (Park et al. 2013), leading to the hypothesis that individuals at risk for cardiovascular disease may present with some type of irisin resistance (Park et al. 2013;Polyzos et al. 2013;Panagiotou et al. 2014).
Exercise has a beneficial effect on cardiovascular function, while sedentary lifestyle is a major risk factor for cardiovascular disease (Joyner and Green 2009;Warren et al. 2010). Regular physical activity is typically associated with lower resting heart rate, which is at least partially attributable to an increase in central cardiac vagal outflow (Melanson 2000;Buchheit et al. 2004;Sandercock et al. 2008;Joyner and Green 2009). The type and intensity of exercise determine the level of resting bradycardia and its mechanisms of control (Azevedo et al. 2014). The current study examined the effect of irisin on cardiac-projecting neurons of nucleus ambiguus, a critical site for central parasympathetic cardiac control (Mendelowitz 1999).

Ethical approval
Animal protocols were approved by the Institutional Animal Care and Use Committees of the Temple University and Thomas Jefferson University. All efforts were made to minimize the number of animals used and their suffering.

Animals
Sprague-Dawley rats (Charles River Laboratories, Wilmington, MA) were used in this study. Neonatal (1-2 days old) rats of either sex were used for retrograde tracing and neuronal culture and adult male rats (250-300 g) were used for the in vivo studies.

Calcium imaging
Measurements of intracellular Ca 2+ concentration, [Ca 2+ ] i were performed as previously described (Brailoiu et al. 2013(Brailoiu et al. , 2014. Briefly, cells were incubated with 5 lmol/L Fura-2 AM (Invitrogen, Life Technologies) in HBSS at room temperature for 45 min, and washed with dye-free HBSS. Coverslips were mounted in an open bath chamber (RP-40LP, Warner Instruments, Hamden, CT) on the stage of an inverted microscope Nikon Eclipse TiE (Nikon Inc., Melville, NY), equipped with a Perfect Focus System and a Photometrics CoolSnap HQ2 CCD camera (Photometrics, Tucson, AZ). During the experiments, the Perfect Focus System was activated. Fura-2 AM fluorescence (emission 510 nm), following alternate excitation at 340 and 380 nm, was acquired at a frequency of 0.25 Hz. Images were acquired/analyzed using NIS-Elements AR software (Nikon). The ratio of the fluorescence signals (340/380 nm) was converted to Ca 2+ concentrations (Grynkiewicz et al. 1985).

Surgical procedures
Adult male Sprague-Dawley rats were anesthetized with a mixture of ketamine hydrochloride (100-150 mg/kg) and acepromazine maleate (0.2 mg/kg) as reported (Brailoiu et al. 2013(Brailoiu et al. , 2014. Animals were placed into a stereotaxic instrument; a guide C315G cannula (PlasticsOne, Roanoke, VA) was bilaterally inserted into the nucleus ambiguus. The stereotaxic coordinates for identification of nucleus ambiguus were: 12.24 mm posterior to bregma, 2.1 mm from midline and 8.2 mm ventral to the dura mater (Praxinos and Watson 1998). A C315DC cannula dummy (PlasticsOne) was used to prevent contamination. For transmitters implantation, a 2-cm long incision was made along the linea alba. A calibrated transmitter (E-mitters, series 4000; Mini-Mitter, Sunriver, OR) was inserted in the intraperitoneal space, as previously described (Brailoiu et al. 2013(Brailoiu et al. , 2014. Subsequently, the abdominal musculature and dermis were sutured independently, and animals returned to individual cages.

Telemetric heart rate monitoring
The signal generated by transmitters was collected via series 4000 receivers (Mini-Mitter, Sunriver, OR), as previously described (Brailoiu et al. 2013(Brailoiu et al. , 2014. VitalView TM software (Mini-Mitter, Sunriver, OR) was used for data acquisition. Each data point represents the average of heart rate per 30 sec.

Noninvasive blood pressure measurement
In rats with cannula inserted into the nucleus ambiguus, blood pressure was noninvasively measured using a volume pressure recording sensor and an occlusion tail-cuff (CODA System, Kent Scientific, Torrington, CT), as described (Brailoiu et al. 2014). One week after the inser-tion of the cannula, rats were exposed to handling and training every day for 1 week. The maximum occlusion pressure was 200 mm Hg, minimum pressure 30 mm Hg and deflation time 10 sec. Two measurements were done per 30 sec (one cycle) and the average was used to calculate heart rate, systolic, diastolic and mean arterial blood pressure. Ten acclimatization cycles were done before starting the experiments.

Microinjection into nucleus ambiguus
One week after surgery (telemetric studies), or after another week of training (tail-cuff measurements), the solution to be tested was bilaterally microinjected into the nucleus ambiguus, using the C315I internal cannula (33 gauge, PlasticsOne), without animal handling. In the tailcuff method, trained rats were in the animal holder for the duration of the experiment. For recovery, at least 2 h were allowed between two injections. Injection of L-glutamate (5 mmol/L, 50 nL with Neuros Hamilton syringe, Model 7000.5 KH SYR) was used for the functional identification of nucleus ambiguus (Brailoiu et al. 2013(Brailoiu et al. , 2014. At the end of the experiments, the microinjection sites were identified, and compared with a standard rat brain atlas (Praxinos and Watson 1998) as previously described (Brailoiu et al. 2013).

Statistical analysis
Data were expressed as mean AE standard error of mean. One-way ANOVA followed by post hoc analysis using Bonferroni and Tukey tests was used to evaluate significant differences between groups; P < 0.05 was considered statistically significant.

A B C D
Cardiovascular monitoring using tail-cuff methods identified similar bradycardic effects (Fig. 3A bottom) to those assessed by telemetry measurement, and absence of any effect on blood pressure (Fig. 3A bottom). As previously reported (Brailoiu et al. 2014), we found a good correlation between these two methods: L-glutamate decreased the heart rate by 89 AE 3.9 bpm (telemetry) and by 87 AE 4.1 bpm (tail cuff), while the bradycardic responses to irisin measured 63 AE 3.4 bpm and 61 AE 3.7 bpm, respectively (N = 5 rats, Fig. 3C). A diagram indicating the sites of microinjection is illustrated in Fig 3D. reported in athletes (Borresen and Lambert 2008). To our knowledge, this is the first report to link a myokine to a cardioprotective role by modulating the central cardiovascular regulation.