Alpha-Klotho is a novel predictor of treatment responsiveness in patients with heart failure

Heart failure is a major cause of death with an increasing population of elderly individuals. Several studies have demonstrated the involvement of soluble alpha-Klotho (sαKl) in various diseases. However, the correlation between sαKl and heart failure remains to be understood. The aim of this study is to investigate the levels and role of sαKl in patients with heart failure. Twenty-eight consecutive patients with acute heart failure (19 male, 9 female), admitted to the Osaka University Hospital from 2010 to 2018, were enrolled in this study. Mean NYHA score, left ventricular ejection fraction and BNP were 3.3, 17.0% and 588 pg/mL, respectively. SαKl significantly increased in heart failure patients. SαKl on admission were significantly higher in patients with heart failure who showed improvement after intensive treatment than that in patients who did not show improvement after the treatment. SαKl levels decreased significantly in patients who showed improvement. Interestingly, sαKl levels increased in male patients with heart failure, but not in female patients. Our data suggest that soluble αKl may be a novel biomarker for the responsiveness against treatment in patients with heart failure with reduced ejection fraction. Our findings may help developing a personalized therapy for different patients with heart failure.

Responsiveness to treatment and sαKl levels. Although all patients received intensive care and medication for heart failure, some patients did not respond well. We classified the patients who showed improvement in their BNP level after the treatment as responders (n = 22) and who did not show improvement as nonresponders (n = 6). As shown in Table 3, serum BNP level significantly improved and EF tended to improve in the responders. In addition, there was a tendency of a decrease in NYHA score in responders compared to non-responders after the treatment.

Gender difference in sαKl.
To evaluate a gender difference in these parameters, we confirmed that there was no significant difference in NYHA (3.4 ± 0.2 vs. 3.3 ± 0.2, P = 0.958), age (44.6 ± 3.0 vs. 50.3 ± 5.1, P = 0.332) or any comorbidity between male and female patients on admission. Then, we analyzed IL-6 levels on admission in the patients, and observed substantial increased levels in both male and female patients (male controls: 1.9 ± 0.1 pg/mL; male patients: 87.6 ± 54.0 pg/mL; female controls: 1.1 ± 0.1 pg/mL; female patients: 14.0 ± 8.4 pg/ mL). We reported that smoking affects serum levels of sαKl differently in males and females 15 . Thus, we next analyzed serum sαKl concentrations on admission of each gender (Fig. 3). There was a significant increase in serum sαKl level in male patients (473.6 ± 39.6 vs. 717.7 ± 58.3 pg/mL, P = 0.0039), however female patients did not show a significant increase (533.4 ± 60.7 vs. 677.8 ± 48.2 pg/mL). Table 3. NYHA score and BNP levels in patients on admission and after treatment. The New York Heart Association functional classification (NYHA) score, serum levels of brain natriuretic peptide (BNP) and left ventricular ejection fraction (EF) were determined in patients on admission and after treatment. The data after treatment were obtained 82.0 ± 13.7 and 43.2 ± 11.1 days after admission in responders and non-responders, respectively (P = 0.173). Data are expressed as mean ± sem. *P < 0.0001 compared to values on admission. † P = 0.0003 compared to values of the responders. Repeated two-way ANOVA followed by Bonferroni's post hoc test was used.  www.nature.com/scientificreports/

Discussion
Recent various studies show an important pathophysiological role of inflammation in the development and progression of heart failure. Serum levels of sαKl change in an age-dependent manner 8 and are associated with anti-inflammatory effects 16 . In this study, we analyzed the correlation between sαKl levels and heart failure. Although serum concentration of sαKl markedly increased in patients with heart failure and decreased upon successful treatment, sαKl levels did not correlate with BNP as well as logBNP (data not shown). This is in accordance with a previous study that shows no correlation between sαKl and severity of heart failure 12 . On www.nature.com/scientificreports/ the other hands, a negative correlation between sαKl and BNP had been reported in the heart failure patients with preserved EF (15). Klotho is predominantly expressed in the kidney and diseased heart tissue 12 , but not in cardiac myocytes 4,17,18 . Although Klotho is a transmembrane protein, its extracellular domain is released from the cell surface and has multiple renal and extrarenal functions. The protective role of sαKl in the murine heart was reported and that Klotho deficiency induced cardiac hypertrophy 18 . Systemically circulating Klotho protects the heart from stress-induced pathological cardiac remodeling and fibrosis 19,20 . Thus, sαKl levels might indicate the status of the heart in patients with heart failure. Serum levels of sαKl markedly increase by smoking and psychological stress in healthy men 10,11 . This increase in sαKl could be attributed to compensation against the negative effects of smoking, such as systemic inflammation. Similarly, there is a possibility that sαKl is produced as a compensatory response and protects the heart during heart failure by acting as a suppressor of inflammation. Thus, patients who respond to stimuli during heart failure and produce sαKl may have a better chance to ameliorate symptoms.

D-Klotho
We evaluated the ratio of IL-6 to sαKl and found that the ratio was higher in non-responder group than in responder group after treatment (1.22 ± 1.08 vs. 0.00974 ± 0.00217, P = 0.0148 analysed by repeated two-way ANOVA followed by Bonferroni's post hoc test), although there was no difference on admission, suggesting that sαKl was not sufficiently produced to regulate inflammation in non-responder group.
The relative risk of developing cardiovascular diseases after smoking is higher in women than that in men, suggesting women have higher sensitivity to the toxic effects of smoking 21 . Men and women have different sαKl responses to stress, such as smoking 15 ; only men exhibit a positive correlation of sαKl levels with smoking status. In this study, similar to the smoking, males showed enhanced sαKl response to heart failure than females. This could be attributed to the poor prognosis of female patients with heart failure. Thus, as compared to male patients, female patients with heart failure require more strict intensive care. However, the non-responders in this study consisted of 5 males and 1 female, suggesting that the reduced responsiveness to treatment may not be only due to the sex of the individual. Furthermore, since female patients with heart failure did not comprise the majority of non-responders, that indicates the presence of other protective mechanisms apart from sαKl in females. On the other hand, it is difficult to be excluded that the smaller sample size (n = 9) could be one of possibilities for the non-significance in serum sα-Kl of female patients.
It is well known that serum level of BNP is a powerful tool for the evaluation of severity and prediction of prognosis for heart failure patients 22 . However, it is still not possible to apply a BNP-guided treatment to all patients with heart failure because of the heterogeneous etiology and patient background. Considering the significance of inflammation in the development of heart failure, sαKl could be an additional factor to determine a prognosis of the disease. sαKl might play a supplemental role to support treatment based on BNP for those patients.
This study is associated with several limitations. First, the study cohort was comprised of a small population; future studies will require an increased number of patients. Second, two out of six non-responders were patients with fulminant myocarditis. Thus, the pathogenesis of heart failure may affect the dynamics of sαKl levels in the non-responders. Third, circulating IL-6 increases during chronic heart failure (CHF) and are higher in patients with severe CHF than that in patients with mild CHF 23 . However, we did not observe a significant difference in the serum levels of IL-6 between heart failure patients and control individuals statistically, although they were markedly increased in male and female patients. This could be because of very low cardiac function and heterogenic etiology of the study subjects. Especially, patients who had fulminant myocarditis were all male. That might be the reason why male patients showed very high level of IL-6 with a large variation.
In conclusion, we have identified sαKl as a novel biomarker for the responsiveness of treatment in heart failure patients with reduced EF. Serum levels of sαKl were different in male and female patients with heart failure. These findings might help develop personalized therapeutic strategies for patients with heart failure.

Methods
Study subjects. The study subjects comprised 28 consecutive patients with acute heart failure who admitted to the Osaka University Hospital from 2010 to 2018 (19 males, 9 females; 24-80 years old). All patients suffered from heart failure with reduced ejection fraction (EF) owing to dilated cardiomyopathy, including idiopathic dilated cardiomyopathy, dilated hypertrophic cardiomyopathy, and ischemic cardiomyopathy (n = 25), and fulminant myocarditis (n = 3). Inclusion criteria were left ventricular EF ≤ 35% and estimated glomerular filtration rate (eGFR) ≥ 30 mL/min/1.73 m 2 (0.739 times for female). It is known that the serum level of BNP is relatively higher and that of soluble alpha-Klotho is lower 24,25 in patients with renal dysfunction compared to healthy controls. To minimize the effect of renal dysfunction on the analyses in this study, we excluded patients with severely reduced eGFR and kidney failure, CKD stage G4 and 5, respectively 26,27 . All patients were intensively treated using catecholamine, left ventricular assist devices, and etc., according to prevailing guidelines. Clinical data, including patient characteristics, laboratory data, and echocardiograms were obtained on admission.
The control subjects were chosen from individuals who underwent health examination at the Osaka University Health and Counseling Center (total 7332 persons). The criteria for the healthy control subjects were no declaration for past disease history, medication, smoking history or any symptom based on health questionnaire. The serum level of soluble alpha-Klotho was measured in randomly selected 115 subjects of the healthy controls. The age-matched controls in this study were randomly selected from the control subjects (19 males, 9 females).
This study was performed in accordance with the ethical guidelines for clinical research of the Ministry of Health, Labor and Welfare and the Ministry of Education, Culture, Sports, Science and Technology of Japan. All the protocols used in this study were approved by the ethics committees of the Osaka University Hospital and Health and Counseling Center, Osaka University. Written informed consent was obtained from all the individuals prior to participation. www.nature.com/scientificreports/