Introduction

Cardiovascular disease is the leading cause of death worldwide (Wang et al. 2016). Even though advancements in biomedical technologies and new treatment modalities have been able to decrease the mortality (but not morbidity) of cardiovascular disease to some extent in the last decade in developed countries, cardiovascular disease continues to put significant burden on the healthcare systems and societies worldwide (Authors/Task Force et al. 2012; Wang et al. 2016). Moreover, the prevalence of cardiovascular disease is expected to rise sharply in developing countries, aggravating the problem (Stewart et al. 2017).

To slow down the progression and development of heart disease, many prevention methods such as life style modifications including a healthy diet, regular exercise and maintenance of optimal blood pressure are promoted. Even though the prevalence of age-related cardiac dysfunction is rising rapidly (Strait and Lakatta 2012), mechanisms shedding light on cardiovascular aging are still not well understood. One of the recently proposed aging theories states that accumulation of damaged macromolecules over lifetime in aging cells may result in the impairment of cellular redox homeostasis, contributing to metabolic dysfunction and aging (Diggs 2008; Gladyshev 2014). One mechanism contributing to cardiovascular aging is free radical induced modification of post-mitotic myocardial macromolecules, which become progressively more prone to oxidative damage during aging process due to impaired redox homeostasis (Cebe et al. 2014b). Our group have shown that supplementation of various chemical substances, such as lipoic acid may have the potential to decrease oxidative modifications in postmitotic tissues (Kayali et al. 2006). The evidence shows that slowing down the formation of oxidatively damaged cellular macromolecules over lifespan may improve cardiovascular disease and extend lifespan (Aydin et al. 2018; Cebe et al. 2014a; Mattison et al. 2017). In this sense, caloric restriction (CR) has been shown to have potential ameliorative effects on lifespan across many species (Liang et al. 2018). On the other hand, many CR studies are met with concern, because it was recently shown that many CR studies are confounded by ad libitum fed rats becoming obese, hence, raising the possibility that the purported improvements in CR groups are due to control (ad libitum fed) rodents becoming obese, rather than there being a real improvement in CR groups (Sohal and Forster 2014). Also, many CR studies remain at an observatory level and only few provide insights into the mechanisms through which CR exerts its effects (Koubova and Guarente 2003).

Therefore, our purpose in this study is two-fold. The first is to investigate the effects of CR on cardiac macromolecules in adult male rats, whose body weight has reached stability. The second, by investigating the continuity of the ameliorative effects of CR after mid-adulthood in our select rat age group, our purpose is to provide a deeper understanding to the molecular mechanisms of CR which could be helpful in the development of drugs that are CR mimetics and do not require dietary restrictions. Hence, in the present study, we decided to assess the effects of CR on the oxidative status of cardiac macromolecules in 6-month-old non-CR rats, 2-year-old CR rats, which were subjected to 40% CR between 18th and 24th months for 6 months, and 2-year-old non-CR rats, which were fed ad libitum, to investigate whether CR started at around middle-ages (18-month-old rat corresponds to 45 human years), could show its favorable redox effects on the metabolomics of heart tissue (Sengupta 2013).

Materials and methods

Chemicals and apparatuses

Supplies were shipped by Sigma-Aldrich (St Louis, MO, USA). The chemicals were stored in a cold room at +4 °C and − 20 °C as described in their manual, and brought to room temperature before the assays. Centrifugation was carried out at +4 °C with a Sigma 3–18 KS centrifuge (SIGMA, Harz, Germany). Redox homeostasis biomarkers of hearts were analyzed by Biotek Synergy™ H1 Hybrid Multi-Mode Microplate Reader (BioTek US, Winooski, VT, USA).

Animal model and treatment protocol

The study was carried out with male Sprague–Dawley rats:

Group I:

Ad libitum fed 6-month-old (n = 7),

Group II:

2-year-old subjected to 40% CR between 18th and 24th months (n = 8),

Group III:

Ad libitum fed 2-year-old (n = 8)

Group II rats were fed ad libitum till 18th months of age; then, their diet was gradually restricted to 40% fewer calories (by providing less rodent chow) than the mean intake of corresponding non-CR rats in Group III. Rats were fed a regular rodent chow, which includes 20% protein, 6% cellulose, and 2% fat. Water was available to all rats ad libitum.

Animal welfare standards were in accordance with the NIH regulations. Rats were kept in a controlled room 21 °C, 50–60% humidity with 12-h light–dark cycles (daytime 07.00–19.00). The study obeys to the local laws and was approved by the Ethics Committee of Bezmialem Vakif University, Istanbul, Turkey. Ethics Committee Issue Number: (2017/236).

Tissue sampling, homogenization and analysis of tissues

After anesthesia with Ketamine HCl; IP, 50 mg/kg; the heart and the roots of great vessels were extracted and washed in cold isotonic saline. The extracted samples were then fresh frozen in liquid N2 until they were homogenized. The samples were homogenized in ice-cold homogenizing buffer (100 mM KH2PO4–K2HPO4, pH 7.4, plus 0.1% (w/v) digitonin) adjusted according to sample weight (Cebe et al. 2014b; Erdogan et al. 2017). After the centrifugation of homogenates at 5000 g for 10 min, redox status biomarkers were assayed with supernatant fraction. The samples were stored for less than 3 weeks at − 80 °C for the investigation of the following redox status biomarkers: protein carbonyl groups (PCO), advanced oxidation protein products (AOPPs), advanced glycation end products (AGEs), soluble receptor for advanced glycation end products (sRAGE), total thiol groups (T-SH), protein thiol groups (P-SH), non-protein thiol groups (NP-SH), dityrosine (DT), kynurenine (KYN), N-formylkynurenine (NFKYN), lipid hydroperoxides (LHPs), 4-hydroxynonenal (4-HNE), malondialdehyde (MDA), total antioxidant status (TAS), catalase (CAT), Cu, Zn-superoxide dismutase (Cu, Zn-SOD)

Analytical methods

Assays of protein oxidation biomarkers

Protein carbonyl groups (PCO)

PCO were assayed as previously described by Reznick and Packer (1994). PCO reacts with 2,4-dinitrophenylhydrazine (DNPH) reagent and forms chromophoric dinitrophenylhydrazones. DNPH modified proteins were precipitated with 20% trichloroacetic acid and pellets were washed with equal amount of ethanol:ethyl acetate mixture for three times. Washing was performed by mechanical disruption of pellets and re-pelleting process was accomplished by centrifugation at 3000 g for 5 min. Finally, the protein precipitates were dissolved in 6 N Guanidine-HCl and the absorbance was recorded at 360 nm.

Advanced oxidation protein products (AOPP)

AOPP were assayed as described by the Hanasand’s method (Hanasand et al. 2012). Supernatant and citric acid were mixed in a microplate. Absorbance was run in duplicate in order to increase the reliability of the AOPP assay at 340 nm and were recorded within 2 min after potassium iodide addition.

Dityrosine (DT)

Protein bound DT was measured as described by spectrofluorometric method of Sadowska-Bartosz et al. (2014). Following the precipitation of proteins, with the help of centrifugation for 15 min at 5000 g the pellets were dissolved and fluorescence spectroscopy was performed. Fluorescence was recorded at 415 nm upon excitation at 330 nm.

Tryptophan derived oxidation markers (KYN, NFKYN)

KYN and NFKYN were measured according to the spectrofluorometric method as described by Sadowska-Bartosz et al. (2014). Following the precipitation of proteins, with the help of centrifugation for 15 min at 5000 g the pellets were dissolved and fluorescence spectroscopy was performed. Fluorescence was recorded at 480, 434 nm upon excitation at 365, 325 nm respectively.

Advanced glycation end products (AGEs) and soluble receptor for advanced glycation end products (sRAGE)

AGE and sRAGE kits were purchased from Shanghai YL Biotech Co. (China). These kits employs sandwich ELISA to measure AGEs and sRAGE. Reagents, samples and standards were prepared according to the instructions of the manufacturer. Sample and buffer solutions were incubated together with specific HRP conjugate in pre-coated plate. Following the incubation period, the wells were washed. The reaction results in a colored complex. Following the reaction, the stop solution is added, and the color of the complex turns into yellow. The spectrophotometric reading was done at 450 nm wavelength.

Assays of lipid peroxidation markers

Lipid hydroperoxides (LHPs)

LHPs levels were assayed spectrophotometrically as described by Wolff SP (Wolff 1994). LHPs oxidize ferrous ions in acid solutions, and the ferric ions are measured by using ferric-sensitive dye, which corresponds to the concentration of LHPs. Xylenol orange binds ferric ions and form a colored (blue-purple) complex. Aliquots of supernatants were then transferred into microcentrifuge tubes. FOX2 reagent (950 μL) was then added. After incubation with FOX2 reagent at ambient temperature for about 30 min, the samples were centrifuged at 3000 g at 20 °C for 10 min and the absorbance values of samples were recorded at 560 nm.

4-Hydroxynonenal (4-HNE)

4-HNE kit was purchased from Shanghai YL Biotech Co. (China). This kit utilizes sandwich ELISA to measure 4-HNE. Reagents, samples and standards were prepared according to the instructions of the manufacturer. Sample and buffer solutions are incubated together with specific HRP conjugate in pre-coated plate. After the reactions, stop solution is added and the color of the complex turns into yellow. The spectrophotometric reading was recorded at 450 nm wavelength.

Malondialdehyde (MDA)

MDA is a late product of lipid peroxidation. MDA reacts with thiobarbituric acid (TBA) to generate a colored product. Lipid peroxidation was determined by the procedure of Buege and Aust (Buege and Aust 1978). The supernatant samples were pretreated to prevent possible interferences according to Lykkesfeldt (2001). The mixture was centrifuged at 3000 g for 5 min and the absorbance of supernatant was recorded at 532 nm. MDA assay was controlled by a duplicate to increase reliability.

Enzymatic and non-enzymatic antioxidant status biomarkers

Total antioxidant status (TAS)

TAS kit was purchased from Shanghai YL Biotech Co. (China). This kit includes IgG antibodies obtained from rabbit and employs sandwich ELISA to measure TAS. Reagents, samples and standards were prepared according to the instructions of the manufacturer. Sample and buffer solutions are incubated together with specific HRP conjugate in a pre-coated plate. The product of the reaction results in a colored complex. Following the reaction, the stop solution was added, which resulted in the complex turning into yellow color. The spectrophotometric reading was recorded at 450 nm wavelength.

Cu, Zn-Superoxide dismutase activity (Cu, Zn-SOD)

Cu, Zn-SOD activity was calculated according to Sun et al. (1988). This method involves the inhibition of nitrobluetetrazolium reduction. The principle of this reaction depends on the utilization of xanthine oxidase (XO) as a superoxide anion generator. At the end of about 20 min of incubation, 50 µL 0.8 mmol/L CuCl2 was added to the microcentrifuge tubes to finalize the reaction. The absorbance was recorded at 560 nm. Inhibition rate: Ablank − Asample/Ablank.100.

Catalase activity (CAT)

CAT activity was calculated according to Aebi (1984). The decomposition of H2O2 was monitored at 240 nm. In this assay, molar absorptivity of 43.6 L mol−1 cm−1 is used to determine the activity of the enzyme.

Thiol fractions (T-SH, P-SH, NP-SH)

Redox sensitive T-SH, P-SH, and NP-SH concentrations of heart tissue were determined by using 5,5-dithiobis (2-nitrobenzoic acid), (DTNB), as described by Sedlak and Lindsay (1968). Supernatant, Tris buffer and DTNB was mixed at pH 8.2 to detect T-SH. NP-SH were assayed by mixing 20 μL of supernatant in 400 μL of 50% TCA. The tubes were centrifuged at 3000 g for 15 min. The absorbance was read at 412 nm.

Data processing

Statistical analysis was performed using IBM SPSS Statistics 24.0. Data were expressed as mean ± standard deviation (SD) for each group. ANOVA with post hoc Bonferroni test was used when ANOVA assumptions were met. As for PCO, DT, AGE, PSH and Cu, Zn-SOD, Kruskal–Wallis H test with Bonferroni correction was used. Two-sided p values are used in the article and p < 0.05 was considered statistically significant.

Results

Table 1 shows the results of redox homeostasis biomarker assays in heart tissue.

Table 1 Redox homeostasis parameters in the male rat heart tissue
Full size table

Figure 1 shows the weight cohort of rat groups.

Fig. 1
figure 1

Line graph cohort which shows the mean weight of the rats. 6-month-age was the end point for the 6-month-old non-CR rats. The weight cohort for the 2-year-old CR and 2-year-old non-CR rats are shown in the figure. As can be seen in the figure, following 6-month-long CR, which was commenced at the 18th month, the weights of CR 2-year-old rats remained comparable to the ad libitum fed rats. CR caloric restriction

Full size image

Assays of protein oxidation

Figure 2 shows the effects of CR on some of the protein oxidation biomarkers (PCO, AOPP and DT).

Fig. 2
figure 2

Scatter plot which shows the effects of CR on protein oxidation biomarkers (PCO, AOPP, DT). Each blue rectangle, green circle and red pentagon represent the concentration of relevant biomarkers obtained from 6-month-old rats (n = 7), CR 2-year-old rats (n = 8) and non-CR 2-year-old rats (n = 8) respectively. AOPP advanced oxidation protein products, CR caloric restriction, DT dityrosine, PCO protein carbonyl groups

Full size image

Protein carbonyl groups (PCO)

Two-year-old CR rats had significantly lower PCO than the corresponding non-CR rats in their heart tissue (p = 0.008). It is worth recognition that whereas 2-year-old CR rats had PCO level that was similar to that of 6-month-old young rats (p = 0.203), 2-year-old ad libitum fed rats had significantly higher PCO level compared to 6-month-old ad libitum fed rats (p = 0.001).

Advanced oxidation protein products (AOPP)

Two-year-old CR rats had significantly lower AOPP than the corresponding non-CR rats in their heart tissue (p = 0.008). While six-month-old ad libitum fed rats had the lowest AOPP level, two-year-old non-CR rats had the highest.

Dityrosine (DT)

Two-year-old CR rats had DT level that was comparable to that of corresponding non-CR rats (p = 0.158). In addition, six-month-old rats had significantly lower DT level than the two-year-old non-CR rats (p < 0.001).

Kynurenine (KYN) and N-Formylkynurenine (NFKYN)

Figure 3 shows the effects of CR on the rest of the analyzed protein oxidation biomarkers KYN and NFKYN. Two-year-old CR rats had significantly lower KYN level than the corresponding non-CR rats in their heart tissue (p < 0.001). In addition, two-year-old CR rats had KYN level that was comparable to that of 6-month-old rats (p = 0.243). On the other hand, two-year-old non-CR rats had significantly higher KYN level than the six-month-old rats (p < 0.001). Hence, six-month-old rats had the lowest and the two-year-old non-CR rats had the highest KYN level.

Fig. 3
figure 3

Scatter plot which shows the effects of CR on protein oxidation biomarkers (KYN, NFKYN). Each blue rectangle, green circle and red pentagon represent the concentration of relevant biomarkers obtained from 6-month-old rats (n = 7), CR 2-year-old rats (n = 8) and non-CR 2-year-old rats (n = 8) respectively. CR caloric restriction, KYN kynurenine, NFKYN N-formylkynurenine

Full size image

Two-year-old CR rats had NFKYN level that was significantly lower than the corresponding non-CR rats (p = 0.004). On the other hand, six-month-old rats also had significantly lower NFKYN level than the two-year-old CR rats (p = 0.033). Thus, six-month-old rats had the lowest and 2-year-old non-CR rats had the highest NFKYN level.

Advanced glycation end products (AGEs) and soluble receptor for advanced glycation end products (sRAGE)

Figure 4 shows the effects of CR on AGEs and sRAGE.

Fig. 4
figure 4

Scatter plot which shows the effects of CR on Advanced glycation end products (AGEs) and Soluble receptor for advanced glycation end products (sRAGE). Each blue rectangle, green circle and red pentagon represent the concentration of relevant biomarkers obtained from 6-month-old rats (n = 7), CR 2-year-old rats (n = 8) and non-CR 2-year-old rats (n = 8) respectively. AGE advanced glycation end products, sRAGE soluble receptor for advanced glycation end products

Full size image

Two-year-old CR rats had significantly lower AGEs than the corresponding non-CR rats in their heart tissue (p = 0.004). It is worth recognition that whereas two-year-old CR rats had AGEs level that was similar to that of six-month-old rats (p = 0.757), two-year-old ad libitum fed rats had significantly higher AGEs level (p < 0.001).

Two-year-old CR rats had significantly lower sRAGEs than the corresponding non-CR rats in their heart tissue (p = 0.027). Overall, the six-month-old control rats had the lowest sRAGEs and two-year-old non-CR rats had the highest.

Assays of lipid oxidation

Figure 5 shows the effects of CR on lipid oxidation biomarkers (LHP, 4-HNE and MDA).

Fig. 5
figure 5

Scatter plot which shows the effects of CR on lipid hydroperoxides (LHPs), 4-hydroxynonenal (4-HNE) and malondialdehyde (MDA). Each blue rectangle, green circle and red pentagon represent the concentration of relevant biomarkers obtained from 6-month-old rats (n = 7), CR 2-year-old rats (n = 8) and non-CR 2-year-old rats (n = 8) respectively. LHP lipid hydroperoxide, 4-HNE 4-hydroxynonenal, MDA malondialdehyde

Full size image

4-Hydroxynonenal (4-HNE)

Two-year-old CR rats had borderline non-significantly lower 4-HNE level than the corresponding non-CR rats in their heart (p = 0.070). In addition, the six-month-old rats had the lowest 4-HNE level compared to the other two groups (p < 0.001).

Lipid hydroperoxides (LHPs)

Two-year-old CR rats had borderline non-significantly lower LHPs than the corresponding non-CR rats in their heart tissue (p = 0.056). In addition, the two-year-old non-CR rats had significantly higher LHPs level than six-month-old rats (p < 0.001).

Malondialdehyde (MDA)

Two-year-old CR rats had MDA level that was comparable to that of corresponding non-CR rats in their heart tissue (p = 1.000). Six-month-old rats had MDA level that was significantly lower than two-year-old non-CR rats (p < 0.001).

Enzymatic and non-enzymatic antioxidant status biomarkers

Figure 6 shows the effects of CR on antioxidative capacity biomarkers (TAS; Cu, Zn-SOD and CAT).

Fig. 6
figure 6

Scatter plot which shows the effects of CR on total antioxidant capacity (TAS); Cu, Zn superoxide dismutase (Cu, Zn-SOD) and catalase (CAT). Each blue rectangle, green circle and red pentagon represent the concentration of relevant biomarkers obtained from 6-month-old rats (n = 7), CR 2-year-old rats (n = 8) and non-CR 2-year-old rats (n = 8) respectively. TAS total antioxidant status, Cu, Zn-SOD superoxide dismutase, CAT catalase

Full size image

Total antioxidant status (TAS)

Two-year-old CR rats had TAS levels that was in between 6-month-old control rats (which had the lowest) and two-year-old ad libitum fed rats (which had the highest). Though, the decrease in the TAS level of two-year-old CR rats was not significant when compared to the corresponding non-CR rats (p = 0.232).

Cu, Zn-Superoxide dismutase activity (Cu, Zn-SOD)

Two-year-old CR rats had significantly higher Cu, Zn-SOD activities than the corresponding non-CR rats in their heart tissue (p = 0.001). While, overall, the six-month-old control rats had the highest Cu, Zn-SOD activities, two-year-old non-CR rats had the lowest.

Catalase (CAT)

Two-year-old CR rats had CAT activity that was comparable to that of corresponding non-CR rats (p = 0.177). Overall, six-month-old rats had the highest CAT activity and the two-year-old non-CR rats had the lowest CAT activity.

Figure 7 shows the effects of CR on antioxidative capacity biomarkers (T-SH, P-SH and NP-SH).

Fig. 7
figure 7

Scatter plot which shows the effects of CR on total thiol (TSH), non-protein thiol (NPSH) and protein thiol (PSH). Each blue rectangle, green circle and red pentagon represent the concentration of relevant biomarkers obtained from 6-month-old rats (n = 7), CR 2-year-old rats (n = 8) and non-CR 2-year-old rats (n = 8) respectively. TSH total thiol, NPSH non-protein thiol, PSH protein thiol

Full size image

Thiol fractions (T-SH, P-SH and NP-SH)

Two-year-old CR rats had significantly higher total thiol groups (T-SH) than the corresponding non-CR rats in their heart tissue (p = 0.008). It is also worth recognition that whereas two-year-old CR rats had T-SH that was similar to that of six-month-old rats (p = 0.237), two-year-old ad libitum fed rats had significantly lower T-SH level compared to six-month-old rats (p < 0.001). Two-year-old CR rats had significantly higher P-SH than the corresponding non-CR rats in their heart tissue (p = 0.008). NP-SH level was comparable in all three groups.

Discussion

In the present study, the effects of CR on the redox status of heart tissue was investigated in six-month-old ad libitum fed rats, two-year-old rats subjected to %40 CR (between 18th and 24th months) and two-year-old ad libitum fed rats. The main purpose of this study was to determine whether the protective effects of CR could be observed in the heart tissue of adult male rats when CR is started around the middle adulthood, when the body weight and composition of weight tends to be stable. Due to the fact that recent reports put the credibility of the ameliorative effects of CR into question, owing to possible confounding factors (such as body weight difference between CR and ad libitum fed rats at end-point of the study and hormonal alterations during female aging), we have decided to use male rats, whose body composition and hormonal status tend to be more consistently reproducible (Sohal and Forster 2014). Even though many effects of CR are seen both in male and female sexes (Colman et al. 1998), some of the previous CR studies indicated that male sex may be at an advantage compared to female sex in some of the biomarkers, such as for lipoprotein (a) (Edwards et al. 2001). When the organism is in net caloric deficit and challenged to spend its energy more efficiently, utilization of previously stored macromolecules will occur. Therefore, the previously observed and relatively different effects of CR on males and females are likely to be due to the body composition of animals when the CR was commenced, rather than inherent differences in CR mechanisms.

Studies focusing on biochemical understanding of aging at the cellular level is scarce. Majority of the previous studies on cardiac aging focus on findings of large scale, such as left ventricular (LV) diastolic function, LV hypertrophy, LV ejection fraction, and cardiac mechanical reserve (Lakatta and Levy 2003; Olivetti et al. 1991; Akasheva et al. 2015). In addition to these findings, previous studies investigating cardiac aging found increased LV fibrosis (Ling et al. 2012) and amyloid deposition (Tanskanen et al. 2008). Our results, which shows increased accumulation of pro-oxidative adducts of macromolecules and depleted antioxidant capacity, as discussed in this section are consistent with these findings and support the anti-aging effects of CR on heart tissue in this age group.

PCO is a commonly used global protein redox status biomarker. Its tissue level elevates in pro-oxidative states and is associated with chronic inflammatory conditions and age-related disorders (Dalle-Donne et al. 2003). In the present study, CR for 6 months significantly improved this parameter compared to corresponding non-CR group. What is also eye opening is, the results indicate that PCO levels of two-year-old CR rats were similar to that of six-month-old rats. This finding is important, because in many studies conducted by our group and others, proved that protein redox status biomarkers (such as PCO), tend to increase during aging (Çakatay et al. 2003; Gryszczyńska et al. 2017). Thus, the fact that in our study 6 month long CR in two-year-old rats diminished PCO levels to the degree that it became comparable to that of 6-month-old rats is especially noteworthy (Fig. 2).

AOPP is another widely used biomarker representing oxidative modifications in a variety of protein molecules. We have previously shown that it increases when the extent of oxidative stress increases in the cellular environment. AOPP is commonly tested as a proxy for macromolecular oxidative alterations and is based on a reliable method (Cebe et al. 2014b). The heart has a high-metabolic demand with ample mitochondria, therefore, it is a high oxidative stress medium and it is likely that the effects of interventions will not probably be diluted (Sawyer 2011). Substantial evidence indicates that pro-oxidative alterations are associated with association of genes involved in cardiac hypertrophy, heart failure and death of myocytes (Siwik et al. 1999). In the present study, six-month-long CR was able to curb the production of AOPP significantly (Fig. 2).

KYN and NFKYN are novel biomarkers associated with chronic inflammation and cardiovascular disease. KYN has been shown to be associated with immune activation and inflammatory response, which plays a role in the etiology of many diseases including cardiovascular disease (Wang et al. 2015). Pawlak K et al. demonstrated that KYN and its derivatives are associated with inflammation and with cardiovascular disease (Pawlak et al. 2009). In the present study, 40% CR has resulted in significantly lower concentrations of KYN (p < 0.001) and NFKYN (p = 0.004) in the heart tissue of two-year-old rats compared to corresponding non-CR rats. In addition, two-year-old rats subjected to CR had KYN level which was similar to (p = 0.243) those of six-month-old rats. Therefore, this parameter as well, supports the idea that CR could render substantial macromolecular advantages. DT structure has been implicated in various inflammatory conditions. Tyrosyl radicals have been shown to play a role in mounting inflammatory response. The increased DT levels are associated with atherogenic plaques, cardiovascular disease and adverse metabolic remodeling (Yang et al. 2017). In our study, two-year-old CR rats had significantly lower (p = 0.010) compared to non-CR rats. Therefore, CR may have substantial effects in the production of DT, which is associated with adverse remodeling (Fig. 2).

AGEs are produced through non-enzymatic glycation and through pro-oxidative modifications of cellular macromolecules. Previous studies indicate that AGEs levels show positive correlation with heart failure and cardiovascular disease (Nenna et al. 2015). AGE level correlates inversely with left ventricular ejection fraction, which is of significant importance in the management of patients (Hegab et al. 2012). AGE has strong relationship with diabetes (Yanar et al. 2018), which has significant cardiovascular complications (Assar et al. 2016; Hartog et al. 2007). In our study, six-month-long CR decreased AGE level significantly (p = 0.001). Therefore, in the light of this evidence, the findings of the present study show that CR may be a part of the wise man’s armamentarium to slow down the progression of age-related cardiovascular disease (Fig. 4).

sRAGE, or soluble RAGE, is a secreted form of receptor for AGE. Even though there are many contradictory reports, sRAGE is proposed as a biomarker for disorders including hypertension, Alzheimer’s disease and cardiovascular disease (Selvin et al. 2013). It is believed that circulating sRAGE competes with AGE to bind RAGE and therefore decreases its unfavorable downstream effects (Geroldi et al. 2006). Thus, it is said that increasing sRAGE may have potential therapeutic effects (Geroldi et al. 2006). On the other hand, other studies indicate that sRAGE level is higher in diabetic patients and is associated with coronary artery disease (Nakamura et al. 2007a, b). In our study, six-month-old CR rats had sRAGE level that was in between 6-month-old rats and two-year-old non-CR rats. On the other hand, two-year-old rats had higher sRAGE level than the corresponding six-month-old rats. Hence, the results of our study imply that the association of increased sRAGE level as a positive prognostic marker is doubtful (Fig. 4).

LHP, 4-HNE and MDA are lipid oxidation biomarkers. Though LHP is one of the early products of lipid oxidation, MDA formation happens in later stages of peroxidation process. 4-HNE as well, is an aldehyde product of lipid oxidation. Our research group have shown that these biomarkers are associated with various age-related diseases ranging from metabolic alterations, such as diabetes (Kayali et al. 2003) to neurodegenerative diseases (Erdogan et al. 2017) to heart disease (Cebe et al. 2014b). In the present study, assessing the lipid biomarkers investigated, CR seems to have mild improvement in LHP (p = 0.056) and 4-HNE (p = 0.070) compared to corresponding non-CR group. On the other hand, MDA level was not found to be different across the groups. This could be due to the fact that CR was started only after the 18th month and was continued for six months. It is possible that, if CR was started earlier in the life period of rats, this may have resulted in more significant alterations (Fig. 5).

TAS; Cu, Zn-SOD and CAT are biomarkers of antioxidative capacity. Antioxidative reserves represent the ability of the tissue to control the inflammatory alterations in the organism. Their depletion has been shown to be associated with many adverse event, such as diabetes, hypertension, atherosclerosis and heart failure (Fukai and Ushio-Fukai 2011). In the present study, CR rats had significantly higher (p < 0.001) Cu, Zn-SOD activity compared to corresponding non-CR rats. Even though there was not significant improvement in TAS and CAT, their concentrations in two-year-old CR rats were in between six-month-old rats and two-year-old non-CR rats, which can be accepted as a favorable response to CR (Fig. 6).

NP-SHs are molecules such as glutathione, CoA, cysteine with free thiol groups in the cell and they are known as free radical scavengers. Thiol groups including protein thiol groups have antioxidant properties in the cell and they work together with other antioxidants, such as glutathione. Therefore, decreased thiol groups are hallmarks of unfavorable oxidative alterations in the cell and are associated with chronic inflammatory conditions. Our previous research has showed that aging myocardial tissue had overall lower macromolecules with thiol groups (Cebe et al. 2014b). In the present study, it is shown that thiol groups are overall significantly increased with CR, and this finding supports our current hypothesis that CR is a potential therapeutic intervention improving cellular redox homeostasis (Fig. 7). In the light of these findings, it is likely that late onset CR, not only through its effects directly on the heart itself, but also by reducing systemic inflammation, oxidative stress, obesity and insulin resistance is likely to improve the systemic health, therefore contribute to improved quality of life and longer lifespan. This is supported by studies conducted on adult non-obese male and female humans, where 2 years of 25% CR resulted in improved general health status, mood, sexual drive, improved sleep quality and decreased mood disturbances (Martin et al. 2016).

In the present study, 16 different redox homeostasis biomarkers are investigated in this study across three groups; namely, six-month-old ad libitum fed rats, two-year-old rats subjected to 40% CR between 18th and 24th months, and their corresponding non-CR controls. Figures 2, 3, 4, 5, 6, 7 indicate that the redox homeostasis parameters of 2-year-old CR rats were improved nearly to the degree that their redox status became comparable to that of 6-month-old rats. Therefore, the results indicate that CR may have significant effects on the aging male heart. It is likely that similar alterations also take place in female hearts. The effects of CR seem to operate through similar mechanisms and be continuous throughout lifespan. CR could ameliorate redox related metabolomic alterations in the aging heart tissue and could delay the age-related unfavorable modifications in the macromolecules, which are associated with cardiovascular disease. Various research groups including our own have been able to reproduce these redox alterations across various organ systems (Aydin et al. 2018; Dalle-Donne et al. 2003; Gryszczyńska et al. 2017). Therefore, it can be concluded that CR is still one of very few ways that has been proven to extend healthy lifespan. It is our understanding that until we understand the cause-effect relationships and unravel the mysteries of aging and come up with CR mimetic drugs, CR might be used as a tough to abide-by, but scientifically proven method to live longer in a subset of patients. In future studies, small animal echocardiographic imaging and in vitro functional assays of cardiomyocytes from various age groups with and without CR could be analyzed to improve our understanding of the mechanisms of aging and CR. These functional studies could also allow us to work on potential therapeutic agents and CR mimetics by generating high quality and reproducible data.

Conclusion

In the present study, we show that effects of CR are distinct from confounding factors which may arise due to uncontrolled weight gain in control groups. Lower protein, lipid oxidation and higher antioxidative enzyme activities observed in the heart tissue of CR group heralds the improvement of myocardial redox homeostasis even though CR is not started in early adulthood.