Beneficial Cardiac Effects of Caloric Restriction Are Lost with Age in a Murine Model of Obesity

  • Majd AlGhatrif
  • Vabren L. Watts
  • Xiaolin Niu
  • Marc Halushka
  • Karen L. Miller
  • Konrad Vandegaer
  • Djahida Bedja
  • Karen Fox-Talbot
  • Alicja Bielawska
  • Kathleen L. Gabrielson
  • Lili A. Barouch

DOI: 10.1007/s12265-013-9453-4

Cite this article as:
AlGhatrif, M., Watts, V.L., Niu, X. et al. J. of Cardiovasc. Trans. Res. (2013) 6: 436. doi:10.1007/s12265-013-9453-4


Obesity is associated with increased diastolic stiffness and myocardial steatosis and dysfunction. The impact of aging on the protective effects of caloric restriction (CR) is not clear. We studied 2-month (younger) and 6–7-month (older)-old ob/ob mice and age-matched C57BL/6J controls (WT). Ob/ob mice were assigned to diet ad libitum or CR for 4 weeks. We performed echocardiograms, myocardial triglyceride assays, Oil Red O staining, and measured free fatty acids, superoxide, NOS activity, ceramide levels, and Western blots. In younger mice, CR restored diastolic function, reversed myocardial steatosis, and upregulated Akt phosphorylation. None of these changes was observed in the older mice; however, CR decreased oxidative stress and normalized NOS activity in these animals. Interestingly, myocardial steatosis was not associated with increased ceramide, but CR altered the composition of ceramides. In this model of obesity, aging attenuates the benefits of CR on myocardial structure and function.


Obesity Caloric restriction Steatosis Lipotoxicity Diastolic dysfunction 


Obesity is associated with many cardiovascular disorders including hypertension, atherosclerosis, heart failure, and in part left ventricular hypertrophy and increased diastolic stiffness [1]. Several studies have shown worsened myocardial hypertrophy and dysfunction in obese leptin-deficient (ob/ob) mice [2, 3, 4] that become more pronounced with age [2, 5].The deleterious cardiac effects in this model of obesity are thought to be due to multiple factors [4] including the increased hemodynamic burden [6], alteration in glucose metabolism and substrate utilization [7], myocardial steatosis [8], and direct effects of leptin deficiency [2].

Specifically, myocardial accumulation of intracellular triglyceride (TG), a process known as cardiac steatosis, has been associated with myocardial dysfunction in ob/ob mice [8]; however, the extent which cardiac steatosis is lipotoxic is still unclear [9, 10]. Myocardial TG accumulation is thought to play a detrimental role via mitochondrial dysfunction [11, 12, 13, 14], which leads to increased reactive oxygen species [15]. In addition, it may result in a buildup of toxic lipid intermediates such as ceramides, leading to increased cardiomyocyte apoptosis [16, 17, 18]. Furthermore, prolonged cardiac steatosis could lead to long-lasting damage such as myocardial fibrosis [9]. On the other hand, TG accumulation could play a protective role by “storing away” toxic lipid intermediates and decreasing the diversion of palmitate into ceramide and other lipotoxic intermediates [19, 20].

Caloric restriction (CR) is known to be a strongly beneficial cardiovascular intervention [21] as it resolves myocardial steatosis and improves cardiac structure and function in several different obesity models [22, 23, 24]. In ob/ob obese mice, the cardiac benefits of CR are less clear. Previous studies have shown that CR in older ob/ob mice neither improves myocardial steatosis and myocardial substrate utilization [25] nor does it downregulate apoptosis-related genes [26]. Interestingly, it reverses myocardial hypertrophy in younger [25] but not older ob/ob mice [2]. The benefits of CR on other myocardial characteristics in this model and the role of aging in this process are not well understood.

In this study, we aimed to examine (1) the impact of aging on the protective effects of caloric restriction and (2) the extent and nature of cardiac lipotoxicity in ob/ob mice treated with caloric restriction. We hypothesized that, due to age-related changes and the prolonged exposure to lipotoxicity, the cardiac benefits of CR would be more pronounced in younger vs. older mice. We also hypothesized that in older obese ob/ob mice caloric restriction might attenuate lipotoxicity and oxidative stress, even though it fails to reverse myocardial steatosis.


Animals and Caloric Restriction Protocol

We studied 2-month (younger) and 6–7-month (older)-old ob/ob mice with C57BL/6J background (n = 2–7 per group) and age-matched C57BL/6J wild-type (WT) controls (n = 3–7 per group) purchased from Jackson Laboratories, Bar Harbor, ME. Ob/ob mice in both age groups were assigned to 4 weeks of unlimited access to standard chow diet ad libitum (AL-ob) vs. caloric restriction (CR-ob) limited to 0.6 g daily, which is the level of restriction needed to induce weight loss at a rate similar to physiological leptin repletion [2, 26, 27].

Mice were housed in an animal facility with a 12-h light–dark cycle and allowed water and food ad libitum except during the period of caloric restriction. After 4 weeks of treatment, mice were euthanized and hearts collected for histological and biochemical lipid quantification, ceramide quantification, NOS activity assay, and lucigenin chemiluminescence test for superoxide levels. The Institutional Animal Care and Use Committee of The Johns Hopkins University School of Medicine approved all protocols and experimental procedures.

Body and Heart Weight Measurements

Animals were weighed at the beginning and at the end of the treatment period. Animals were euthanized and hearts collected and weighed after removal of pericardial fat. Tibia lengths were measured. Heart weight was normalized to tibia length and reported as normalized heart weight = heart weight/tibia length (milligrams per millimeter).


We performed echocardiography on conscious mice after 4 weeks of treatment as described previously [2]. Wall thickness and diastolic and systolic left ventricular (LV) dimensions were recorded from M-mode images using averaged measurements from three to five consecutive cardiac cycles. LV mass was calculated and indexed to tibia length using the following equation:

Normalized LV mass = 1.055 × [(IVSD + LVEDD + PWT)3 − LVEDD3]/tibia length, where 1.055 is the specific gravity of myocardium; IVSD, interventricular septal thickness at end diastole; PWT, posterior wall thickness at end diastole; and LVEDD, left ventricular end-diastolic diameter. IVSD and PWT were averaged to calculate mean ventricular wall thickness (VWT). Systolic function was assessed by fractional shortening (FS) which was calculated as: FS = 100 % × (LVEDD − LVESD)/LVEDD. Diastolic function was assessed by using Tissue Doppler to assess isovolumetric relaxation time (IVRT), one of the most stable markers of diastolic dysfunction in obesity [28]. IVRT is measured as the time interval from the aortic valve closure at the end of the LV outflow envelope to the mitral valve opening at the beginning of the mitral E wave.

Myocardial Lipids

We performed myocardial lipid measurement on younger mice as we previously described [26]. Since we have reported the results on older mice in a prior publication they are not repeated here [26].

Oil Red O Staining

As previously described [26], myocardial frozen sections of 10 μm were air dried for 30 min and then fixed in 10 % phosphate-buffered formalin for 10 min. After two rinses in distilled water, the sections were stained using an Oil Red O kit (Poly Scientific R & D, Bay Shore, NY). The results were scored semiquantitatively (0 = none, 3 = severe) by a pathologist blinded to the assigned groups.


As previously described [26], measurements were done using an Adipogenesis Assay Kit (Biovision, K610-100). Flash frozen portions of the left ventricle were weighed and homogenized in 10× volume lipid extraction solution provided in the kit. Homogenates were heated at 95 °C for 30 min, vortexed for 1 min, and centrifuged at 12 K for 10 min to remove debris/insoluble material. Supernatants were transferred to a separate tube, volumes were measured, and samples were assayed in duplicate +/− lipase to account for endogenous glycerol in the tissue. For subsequent protein quantification, the insoluble pellet was dried, and then dissolved in 5 % NaOH overnight. Triglyceride (TG) was quantified using colorimetric assay and normalized to milligrams of protein.

Free Fatty Acids

Free fatty acids (FFA) were measured using a Free Fatty Acid Quantification Kit (Biovision, K612-100) as previously described [26]. Flash frozen portions of the left ventricle were weighed and homogenized in 10× volume of chloroform/Triton X-100 (1 % triton in chloroform). The extracts were centrifuged for 10 min at 12 K in a microcentrifuge. The organic phase was transferred to another tube and air dried at 50 °C to remove chloroform. The samples were then vacuum-dried for 30 min to remove the last of the chloroform. The dried lipids were resuspended in 120 μl of Fatty Acid Assay Buffer by vortexing extensively for 5 min. Duplicate 50-μl aliquots were measured. For subsequent protein quantification, the insoluble pellet from the extraction was air dried then dissolved in 5 % NaOH overnight. FFA was quantified using colorimetric assay and normalized to milligrams of protein.

Measurement of Myocardial Superoxide

Myocardial superoxide was measured by lucigenin-enhanced chemiluminescence. Fresh-frozen myocardium was homogenized in 20 mM HEPES buffer and diluted in lucigenin buffer. Lucigenin was used at a sufficiently low concentration (5 μmol/L) that its own effect on the measured superoxide level was negligible [29]. Superoxide-generated signals were recorded as count per minute by a liquid scintillation counter (LS6000IC, Beckman Instruments, Fullerton, CA) as previously described [30].

Measurement of Cardiac NOS Activity

NOS calcium-dependent activity was assessed from myocardial homogenates by measuring C14 arginine to citrulline conversion (assay kits from Stratagene, La Jolla, CA or Cayman Chemical, Ann Arbor, MI) as previously described [31].

Myocardial Ceramide Assay

Liquid chromatography tandem mass spectrometry (LC-MS/MS) analysis of cellular sphingolopids was conducted on younger and older mice. Snap-frozen LV tissue was homogenized in cell lysis buffer as previously described [5]. Advanced analysis of ceramide species was performed by the Lipidomics Shared Resources at the Medical University of South Carolina, Charleston, SC on a ThermoFinnigan TSQ 7000, triple-stage quadrupole mass spectrometer operating in a Multiple Reaction Monitoring positive ionization modes previously described [32]. Final results were expressed as the level of the particular SPLs/protein determined from a colorimetric protocol on the dissolved pellets and expressed as SPLs/Protein (picomoles per milligram) [32].

Western Blot Analysis

Snap-frozen LV tissue was homogenized in cell lysis buffer (Cell Signaling Technology, Danvers, MA) and subjected to Western blot analysis as previously described [5]. Primary rabbit antibodies to total Akt (tAkt) and phosphorylated Akt at Ser473 (pAkt-Ser473) were used at dilution of 1:1,000 (Cell Signaling Technology, Beverly, MA). Secondary goat/anti-rabbit antibodies were used at dilution of 1:10,000 (Cell Signaling Technology, Beverly, MA). The densitometric volume of digitalized band was evaluated by Image J program. Phosphorylation of Akt (pAkt) was presented as pAkt/tAkt.

Statistical Analysis

All data are expressed as mean ± standard error of the mean (SE). Group data were compared using one-way ANOVA with a Tukey’s post hoc test for multiple comparisons. P values ≤ 0.05 were considered to be statistically significant. GraphPad Prism 5.0 (La Jolla, CA) was used for graph production and statistical analysis.


CR Induces Reduction in Heart Weight in Younger but Not Older Mice

CR reduced body weight to a similar extent in both older and younger mice at the end of the 4 weeks (Fig. 1a, b). In younger but not older mice, CR reversed the increased heart weight observed in AL-ob compared to WT mice (Fig.1c).
Fig. 1

a Body weight at baseline. b Body weight after 4 weeks of treatment. c Heart weight normalized to tibia length. Data expressed as mean ± SE. *P < 0.05 vs. WT of the same age. P value vs. younger mice. WT wild-type mice, AL-ob ad libitum fed ob/ob mice, CR-ob caloric-restricted ob/ob mice

CR Improves Diastolic Function in Younger Mice, but This Effect Is Lost with in Older Mice

Echocardiography after 4 weeks of treatment showed that younger and older AL-ob mice had increased LV chamber dilation (Fig. 2a, b), increased normalized LV mass (Fig. 2c), and worse diastolic function measured by isovolumetric relaxation time (IVRT; Fig. 2e).Younger AL-ob mice also had increased wall thickness (VWT; Fig. 2d) and depressed systolic function (Fig. 2F) compared to their WT counterparts, but these did not reach statistical significance in older mice. As a result, VWT was higher in older vs. younger WT mice; however, this age difference was not present in AL-ob. CR normalized all parameters in younger ob/ob mice. Importantly, although CR improved LV chamber dilation and reduced indexed LV mass in older ob/ob mice, it completely failed to improve diastolic function (Fig. 2f, g).
Fig. 2

Echocardiography at 4 weeks of treatment. a Left ventricular end systolic diameter (LVESD). b Left ventricular end diastolic diameter (LVEDD). c Calculated LV mass. d Ventricular wall thickness (VWT) as the average of measured posterior wall thickness and interventricular septal thickness. e Fractional shortening (FS) calculated as FS = 100 % × (LVEDD − LVESD)/LVEDD. f Isovolumetric relaxation time (IVRT). g Representative left ventricular outflow track flow velocity waves illustrating IVRT; dotted line indicates aortic valve closure, and dashed line indicates beginning of E wave. Data expressed as mean ± SE. *P < 0.05 vs. WT of the same age. P value vs. younger mice. WT wild-type mice, Al-ob ad libitum fed ob/ob mice, CR-ob caloric-restricted ob/ob mice

CR Improved Myocardial Steatosis in Younger Mice

Oil red O staining and myocardial TG quantification showed an increase in myocardial lipid content in AL-ob vs. WT mice, which was reversed with CR (Fig. 3a–c). FFA levels (Fig. 3d) showed a trend towards lower FFA in AL-ob mice that was normalized with CR; however, these changes did not reach statistical significance. In contrast, we have previously shown that the same degree of CR did not reverse myocardial steatosis in older mice [26].
Fig. 3

Myocardial lipid content in younger mice a Representative sample of Oil red O staining of myocardial tissue showing deeper staining in AL-ob vs. WT which was improved by CR. b Semiquantitative scoring of Oil Red O staining of tissue from the left ventricle. c Myocardial triglycerides (TG) levels in myocardial tissue. d Free fatty acid (FFA) levels in myocardial tissue. Data expressed as mean ± SE. *P < 0.05 vs. all other groups. WT wild-type mice, AL-ob ad libitum fed ob/ob mice, CR-ob caloric-restricted ob/ob mice

Myocardial Steatosis Was Associated with Increased Oxidative Stress and Upregulation in NOS Activity that Both Reversed with CR

Lucigenin-enhanced chemiluminescence showed increased superoxide in older AL-ob mice that, surprisingly, was reduced by CR (Fig. 4a). Moreover, older AL-ob mice had a significant compensatory increase in NOS activity which was also normalized with CR (Fig. 4b).
Fig. 4

a Myocardial superoxide assessed by lucigenin-enhanced chemiluminescence assay b Myocardial NOS calcium-dependent activity measured by C14 arginine to citrulline conversion. Data expressed as mean ± SE. *P < 0.05 vs. all other groups. WT wild-type mice, AL-ob ad libitum fed ob/ob mice, CR-ob caloric-restricted ob/ob mice

CR Changed Ceramide Composition but Did Not Alter Total Ceramide

We measured ceramide levels in LV tissue of younger and older mice using LC-MS/MS (Fig. 5). Surprisingly, neither total ceramide nor its major subgroups of long chain and very long ceramides levels differed between the mice groups. However, when we looked at separate acyl chain lengths, we found that in older mice, CR had lower levels of C20 and C22 compared to WT and AL-ob. Interestingly, the changes induced by CR were in opposite direction of the changes expected with aging.
Fig. 5

Myocardial ceramide in younger and older mice presented as total, and acyl chain length-specific ceramides. Acyl chain lengths between C14 and C26 were detected and categorized as long chains (C14-20) and very long chains (C22-26). Levels of the main acyl chain length C18-C22:1 are reported separately. Data expressed as mean ± SE.*P < 0.05 vs. all other groups of the same age. WT wild-type mice, AL-ob ad libitum fed ob/ob mice, CR-ob caloric-restricted ob/ob mice

CR Was Associated with Upregulation of Myocardial Akt Phosphorylation in Younger but Not Older Mice Indicating Better Myocardial

We found that Akt phosphorylation did not change with obesity in both younger and older mice. However, CR significantly upregulated Akt phosphorylation in younger but not older mice. Of note, Akt phosphorylation was elevated in older WT and AL-ob compared to their younger counterparts (Fig. 6).
Fig. 6

Western blot analysis for total Akt (tAkt) and Phospho-Akt (Ser473) (pAkt) in young and older mice. Data expressed as mean ± SE of pAkt/tAkt. *P < 0.05 vs. all other groups. WT wild-type mice, AL-ob ad libitum fed ob/ob mice, CR-ob caloric-restricted ob/ob mice


The major new finding of this study is that CR restored normal diastolic function in younger but not older obese ob/ob mice, with a parallel improvement in myocardial steatosis. Although CR did not reverse myocardial steatosis in older ob/ob mice, it did reduce oxidative stress, indicating attenuation in myocardial lipotoxicity. Changes in ceramide did not appear to have a major role in myocardial lipotoxicity in this model; however, CR altered ceramide composition to some extent in an anti-aging pattern. The cardiac benefits observed in younger ob/ob mice were associated with upregulation of Akt phosphorylation.

Caloric Restriction Improves Diastolic Function in Parallel with Improvement in Myocardial Steatosis in Younger but Not Older Obese Mice

We reported worsened diastolic function and myocardial steatosis in both younger and older obese mice. These changes were reversed with CR in younger mice only. These findings are consistent with previous reports showing a direct association between myocardial lipid accumulation and diastolic dysfunction [8, 24, 33, 34]. It is striking that the ability to improve diastolic function is lost with age in this model. We also found that younger mice had diminished fractional shortening (FS) that was reversed with CR.

Our findings of improved myocardial steatosis in younger leptin-deficient mice concur with our previous report of improved overall metabolic profile of younger mice with caloric restriction, reflected by decreased serum glucose and triglyceride levels [35]. However, these results come in contrast to a previous report showing failure to reverse myocardial steatosis with CR [25]. It is important to highlight that in our study CR-ob received 30 % of caloric intake of that of AL-ob compared to approximately 70 % in the study mentioned above [25]. It is likely that the stricter diet we used in our study accounts for this difference in cardiac effects. Of note, this caloric restriction regimen was adopted from our previous experiments in which caloric restriction was titrated to induce weight loss at a rate similar to physiological leptin repletion [2, 26, 27]. Following the same CR regimen assures comparability with our previous studies on older mice by inducing similar rates of weight loss.

Obesity Induces Concentric LVH in Younger Mice, but Does Not Worsen That Observed with Aging

Obesity is associated with LVH in both younger and older mouse models. Older animals have shown a pattern more consistent with eccentric LVH, as demonstrated by increased LV chamber dimensions, possibly in part due to volume expansion [4, 36]. In our current model, younger obese animals show primarily concentric LVH, only in younger mice, reflected by increased VWT and heart weight. Surprisingly, obesity in older mice did not impose any further increase in VWT or heart weight beyond that of age. This could be due to the dominant effect of aging on concentric LVH [37]. A lack of additive effect between obesity and aging has been shown in other aspects of myocardial function [38]. It is worth noting that calculated LV mass showed similar changes in younger and older mice; however, we believe the changes observed are driven more by chamber volume rather muscle mass changes since these changes were not reflected in actual heart weights.

CR Reduced Oxidative Stress Despite Persistent Myocardial Steatosis in Older Mice

In this paper, we have shown that CR did not improve diastolic function in older obese mice, which is consistent with the failure to reduce myocardial steatosis we found previously [26]. However, since we have previously observed improved mitochondrial coupling in older CR-ob mice [26], we wondered whether CR attenuated lipotoxicity in older mice despite persistent steatosis. We did not evaluate the effect of CR on lipotoxicity in younger mice since we already know that CR improves myocardial steatosis. We found that older AL-ob mice experienced not only increased oxidative stress but also upregulation in NOS activity, both of which were reversed by CR. Given the compensatory and protective increase in NOS activity, these findings suggest that mitochondrial uncoupling is a possible source for superoxide buildup in AL-ob mice, and not NOS uncoupling, which would have resulted in direct production of superoxide by NOS itself and a decrease in NOS activity [39].

CR modifies the lipotoxic profile of myocardial steatosis by improving mitochondrial coupling through mechanisms that are not clear. The fact that attenuation of lipotoxicity was not paralleled with improvement in diastolic function in older mice is probably due to irreversible accumulative oxidative damage over time [40]. We have previously shown that obese leptin-deficient mice experience increased DNA damage and higher rates of apoptosis that are accentuated with aging [5]. We have also found that CR, despite restoring mitochondrial function, fails to normalize the expression associated with oxidative damage and cell death in obese leptin-deficient older mice, indicating irreversible damage. Further studies are needed to assess whether CR in younger mice reduces DNA damage and apoptosis, similar to its effects on myocardial steatosis and diastolic dysfunction.

Irreversible cumulative oxidative damage has been observed in many tissues and is thought to be a major player in aging [41]. The interaction between age- and disease-associated oxidative stress is implicated in cardiovascular aging in particular [42]. Furthermore, an age–disease interaction has been observed in other aging processes; for example, advanced glycation, a process associated with both aging and diabetes, results in time-dependent irreversible changes in cellular and physicochemical properties of tissues [43].

CR Changes Ceramide Composition, Although Neither Ceramide nor Its Composition Changed with Obesity

We measured myocardial ceramide level to investigate other aspects of lipotoxicity and their alteration with CR. Surprisingly, myocardial steatosis in younger and older obese mice was not associated with an increase in total ceramide or its composition, suggesting that myocardial steatosis in AL-ob may lead to sequestration of excess ceramide [19, 44]. Interestingly, CR decreased the levels of C20 and C22, major ceramide acyl chains, in the opposite direction to that observed with aging. Similarly, a previous study examining renal ceramide reported an anti-aging effect of CR on other ceramide species [45]. The role of the different chain lengths of ceramide in cardiac tissue is not fully understood and warrants further study.

Age Difference in the Effect of CR on Myocardial Steatosis Was Explained by Increased Myocardial pAkt

We next examined the possibility of improvement in myocardial glucose metabolism and insulin sensitivity with CR in younger but not older mice. Our findings of increased pAkt-Ser473, a downstream insulin signaling pathway, with CR in younger but not older mice support this hypothesis. Interestingly, pAkt-Ser473 was elevated across the older groups. A previous study has shown that increased pAkt-Ser473 with aging is a marker of Akt dysfunction with diminished rather than increased insulin response in skeletal muscles [46]. These findings suggest that age-associated Akt dysfunction and subsequent saturation of Akt phosphorylation might be a mechanism behind persistent steatosis in older mice treated with CR. Akt phosphorylation increases despite decreased Akt kinase activity and downregulation of upstream molecules [46]. Hence, Akt hyperphosphorylation is potentially due to decreased phosphatase activity rather than increased kinase activity. Future studies are needed to investigate whether Akt phosphatase enzymes such as protein phosphatase 2A are impaired with aging.


In this model, obesity-associated worsening of diastolic function is independent of age. Although diastolic function is associated with myocardial steatosis in younger and older obese mice, improvements with CR were apparent in younger mice only. This protective effect was completely lost in the older group. The improvement observed with caloric restriction in younger mice was associated with upregulation in Akt phosphorylation, a change not observed in older mice. This suggests that age-associated Akt dysfunction might play a role in the loss of protective effects with age. Interestingly, although CR in older mice does not resolve myocardial steatosis, it did attenuate oxidative stress and lipotoxicity in these animals. We therefore speculate that the prolonged exposure to accumulated oxidative damage may explain the lack of improvement in myocardial function in older mice despite the acute reduction of oxidative stress. Additional studies are needed to determine the exact mechanism involved and the point at which oxidative damage no longer allows reversal of diastolic function abnormalities with caloric restriction.

In conclusion, we have uncovered an age-dependent pathway leading to obesity-related diastolic dysfunction that is only reversible in younger animals. Discovering the exact mechanism of reversibility and determining what ultimately can reduce the persistent oxidative damage and diastolic dysfunction will greatly enhance our understanding of the physiology of cardiovascular aging.


The authors are grateful for the financial support of the American Heart Association Beginning Grant-In-Aid [to L.A.B.], American Diabetes Association [to L.A.B.], and the National Institutes of Health [5T32HL007227 to V.L.W]. There are no relationships to disclose.

Copyright information

© Springer Science+Business Media New York 2013

Authors and Affiliations

  • Majd AlGhatrif
    • 1
    • 2
  • Vabren L. Watts
    • 1
  • Xiaolin Niu
    • 1
    • 3
  • Marc Halushka
    • 5
  • Karen L. Miller
    • 1
  • Konrad Vandegaer
    • 1
  • Djahida Bedja
    • 4
  • Karen Fox-Talbot
    • 5
  • Alicja Bielawska
    • 6
  • Kathleen L. Gabrielson
    • 4
  • Lili A. Barouch
    • 1
  1. 1.Division of Cardiology, Department of MedicineJohns Hopkins University School of MedicineBaltimoreUSA
  2. 2.Division of Hospital Medicine, Bayview Department of MedicineJohns Hopkins University School of MedicineBaltimoreUSA
  3. 3.Department of Cardiology, Tangdu HospitalThe Fourth Military Medical UniversityXi’anPeople’s Republic of China
  4. 4.Department of Comparative MedicineJohns Hopkins University School of MedicineBaltimoreUSA
  5. 5.Department of PathologyJohns Hopkins University School of MedicineBaltimoreUSA
  6. 6.Department of Biochemistry and Molecular BiologyMedical University of South CarolinaCharlestonUSA

Personalised recommendations