OUP user menu

Postischemic myocardial recovery and oxidative stress status of vitamin C deficient rat hearts

Catherine Vergely, Caroline Perrin, Aline Laubriet, Alexandra Oudot, Marianne Zeller, Jean-Claude Guilland, Luc Rochette
DOI: http://dx.doi.org/10.1016/S0008-6363(01)00247-4 89-99 First published online: 1 July 2001


Objective: To investigate the role of vitamin C tissue content as a protective agent during myocardial ischemia–reperfusion injury, we have evaluated the postischemic functional recovery and free radical release of osteogenic disorder Shionogi (ODS) inherently scorbutic rat hearts and compared them to healthy Wistar rat hearts. Methods: Isolated perfused hearts of ODS or Wistar rats underwent 30 min of a global total normothermic ischemia followed by 30 min of reperfusion. The lipid-soluble spin trap α-phenyl N-tert-butylnitrone (3 mM) was perfused upstream of the coronary bed. Functional parameters were recorded and samples of coronary effluents were analysed using electron spin resonance spectroscopy to characterise and quantify the amount of radical species released. Results: From the onset of reperfusion, a large and long-lasting release of alkyl/alkoxyl radicals was detected, with a peak value of 29.0±3.2 nM obtained after 13 min, which was associated with a persistent contractile dysfunction. However, ODS rat hearts showed a higher myocardial recovery with lower left ventricular end diastolic pressure (44.34±1.74 vs. 55.03±1.57 mmHg for Wistar), higher recovery of rate pressure product (12.3±1.4 vs. 1.9±1.7×103 mmHgbeats/min for Wistar) and shorter duration of contractile abnormalities during reperfusion (3.7±1.0 vs. 20.8±5.3 min for Wistar). Moreover, free radical release was identical in ODS rat hearts as compared to control Wistar rats. Ascorbic acid tissue content was significantly altered in ODS rats (31.9±3.3 vs. 591.0±54.9 mmol/g of tissue for Wistar) but superoxide dismutases, glutathion peroxidases and inducible heat shock protein 70 genes were up-regulated. Conclusions: This study shows that ascorbic-acid-deficient ODS rat hearts are more resistant to an ischemic insult than control Wistar rats, probably through the development of alternative protective defences, like the induction of heat shock proteins. These paradoxical results raise the question of the relative importance of each endogenous antioxidant in the cardiac resistance to ischemia–reperfusion injury.

  • Free radicals
  • Gene expression
  • Ischemia
  • NMR
  • Reperfusion
  • Ventricular function

Time for primary review 32 days.

1 Introduction

Cellular damage by reactive oxygen species such as superoxide or hydroxyl radicals is generally believed to be a significant causal factor involved in heart diseases, especially during myocardial ischemia–reperfusion. Many authors have provided extensive evidence that free radicals are produced and released from the ischemic heart and, to a larger extent, during the reperfusion period [1–4]. The oxidative stress associated with ischemia–reperfusion is thought to be closely related to the impairment of antioxidant levels in ischemic tissues [5,6] followed by a burst in free radical generating system production, and is tightly linked to the impairment of functional recovery [3].

The modes of action of many of the natural antioxidants that are found in biological fluids and tissues have been the subject of intense investigation. In vivo, the relative importance of each of the antioxidant agents and their individual contribution to the total antioxidant capacity of a biological system is still, however, not clearly understood. Ascorbic acid is widely distributed in cells and is regarded as the most potent water-soluble antioxidant found in plasma and biological fluids [7]. However, most of its antioxidant properties have been evaluated in vitro and the biological relevance of these experiments is difficult to establish in vivo [8].

Therefore, to investigate the role of ascorbic acid in vivo, we have used a mutant strain of Wistar rats with a hereditary osteogenic disorder [9] which involves a deficiency of l-gulonolactone oxidase [10] a key enzyme catalysing the last step of the hepatic pathway from d-glucose to l-ascorbic acid. The lack of the enzyme is due to a single base mutation from G to A at nucleotide 182, a mutation that alters the 61st amino acid residue from Cys to Tyr [11]. The disorder is controlled by a single autosomal recessive gene and homozygotes of the strain (od/od) which are named osteogenic disorder syndrome (ODS) Shionogi rats manifest the syndrome. Like humans, other primates and guinea pigs, these mutants become scorbutic without a supply of ascorbic acid.

Therefore, to improve our comprehension of the relationship between tissue antioxidant status and the response to an ischemic insult, we have aimed our study to evaluate the postischemic recovery and the free radical release of vitamin C deficient ODS rat hearts and to compare them with healthy control Wistar rat hearts.

2 Methods

2.1 Chemicals

All chemicals were purchased from Sigma (France). The spin trap α-phenyl-N-tert-butylnitrone (PBN) was purified through sublimation under argon gas.

2.2 Rats and diets

The investigators conforms with the authorisation 00775 from the French government which agrees with the Guide for the Care and Use of Laboratory Animals published by the US National Institutes of Health. ODS male rats aged 8 months old were purchased from Clea Laboratories (Japan). The rats were fed a purified diet (Inra-Upae, Centre de Recherche de Jouy, Jouy-en-Josas, France) to which a known amount of l-ascorbic acid had been added. The basic diet, contained 65% of total energy as carbohydrate, 23.4% as casein and 11.6% as vegetable fat. Two diets were used: a vitamin-C-deficient diet and a vitamin-C-supplemented diet, formulated to provide 0 mg and 800 mg/kg, respectively. Two weeks before the beginning of the experiment with ODS rats, the vitamin-C-supplemented diet was changed to a vitamin-C-deficient diet.

Wistar rats aged 8 months old were purchased from Depré (France) and fed with the purified vitamin-C-deficient diet. The US National Research Council considers that normal rats do not require a dietary source of ascorbic acid.

2.3 Perfusion technique and perfusion medium

The rats were anaesthetised with sodium thiopental (60 mg per kg, i.p.) and heparin was intravenously injected (500 I.U./kg). After 1 min, the hearts were excised and placed in a cold (4°C) perfusion buffer bath until contractions ceased. Each heart was then immediately cannulated through the aorta and perfused by the Langendorff method, at a constant perfusion pressure equivalent to 80 cm of water (8 kPa). The perfusion buffer consisted of a modified Krebs–Henseleit bicarbonate buffer (mM concentrations: NaCl 118, NaHCO3 25, MgSO4 1.2, KH2PO4 1.2, KCl 4.7, glucose 5.5 and CaCl2 3.0). Before use, all solutions were filtered through an 0.8 μm Millipore filter to remove any particulate contaminants. The perfusion fluid was gassed with 95% oxygen–5% carbon dioxide (pH 7.3–7.5 at 37°C). An elastic water-filled latex balloon (no. 4, Hugo Sachs, Germany) was inserted into the left ventricle through the mitral valve and connected to a pressure transducer, the output of which was connected to a physiograph. The filling pressure was individually adjusted to 12–18 mmHg (1.6–2.5 kPa) to achieve a maximal contractile performance. A Gould TA 240 recorder was used to measure intraventricular pressures and heart rate. Coronary flow was measured by the timed collection of the effluent. The spin trap α-phenyl-N-tert-butylnitrone (PBN) was dissolved at a concentration of 120 mM in NaCl 0.9% gassed with N2 to exclude oxygen. PBN was administered upstream of the coronary bed with a mini pump (Harvard Apparatus), with an infusion rate adjusted to 1/40th of the coronary flow, ensuring a final PBN concentration of 3 mM. Samples of coronary effluents were collected at various times for further determinations of spin adduct levels with electron spin resonance (ESR) spectroscopy. All procedures using PBN were performed away from light and oxygen to avoid light-decomposition and oxidation of the spin trap.

2.4 Perfusion protocols

Four groups, each composed of six hearts, were subjected to different ischemia–reperfusion protocols at 37°C (Fig. 1). After a stabilisation phase of 15 min, isolated hearts were perfused aerobically for 15 min (preischemic control period). Global normothermic ischemia was then induced by clamping aortic inflow for 30 min, during which a thermoregulated chamber maintained the heart temperature at 37°C. After ischemia, aortic inflow was resumed for 30 min (reperfusion period) which is the usual duration of reperfusion chosen to study the functional effects of reperfusion associated with oxidative stress.

Fig. 1

Perfusion protocols of Wistar or ODS ascorbate-deficient rat hearts. Arrows (↑) indicate time for coronary sample collection.

PBN or vehicle was infused 5 min before the onset of ischemia and during the reperfusion period (15 min from the beginning, and 5 min before the end of the reperfusion period). Aliquots (5 ml) of coronary effluent samples were collected at different times before ischemia and during reperfusion (see arrows in Fig. 1), immediately extracted with N2-gassed ice-cold toluene (0.75 ml), frozen and kept in liquid nitrogen until ESR measurement.

2.5 ESR spin trapping

Toluene extracts were thawed and bubbled with N2 for 20 s. ESR spectra were recorded at 293 K with a Bruker ESP 300E-X band spectrometer using a TM110 cavity and an aqueous flat cell. The following parameters were selected for optimal detection of PBN spin adducts in coronary effluents: microwave power, 20 mW; microwave frequency, 9.74 GHz; modulation amplitude, 1.6 G; modulation frequency, 100 kHz; gain, 1.6–3.2·106; scan rate, 0.95 G·s−1; time constant, 163.84 ms; conversion time, 82 ms.

The signal intensity which is proportional to the concentration of spin adducts was measured directly from the field scan and expressed as spin adduct concentration (nM) by double integration of the experimental spectra using 2,2,6,6-tetramethylpiperidine-N-oxyl (TEMPO) nitroxide as an integration standard. The amount of myocardial PBN spin adduct liberation (pmol/min/g of heart) at each perfusion time was obtained by multiplying the adduct concentration by the respective coronary flow.

2.6 Plasma and tissue processing

After anaesthesia, the blood of the rats was collected from the abdominal aorta, centrifuged at 4°C and the plasma was aliquoted and stored at −70°C until analysis. For vitamin C measurements, 2 volumes of metaphosphoric acid solution at 5% were added to plasma samples before freezing and storage.

After each experimental ischemia–reperfusion protocol, atria were rapidly excised and the remainder of the heart was instantaneously frozen and crushed in liquid nitrogen and kept at −70°C until use. Only control Wistar and ODS rat hearts untreated with PBN were kept for biochemical analyses. Afterwards, hearts were homogenised in 5 volumes of either 50 mM phosphate buffer (pH 7.4, 0.1 mM EDTA) or 0.25 M sucrose solution, or Tri-reagent. Phosphate homogenates were used for vitamin C, vitamin E, lipid peroxidation and protein determinations. For vitamin C, 2 volumes of metaphosphoric acid solution at 5% were added to samples before freezing and storage. Sucrose homogenates were centrifuged at 6500 g at 4°C, and the supernatant was then used to evaluate enzymatic activities. Tri-reagent was used to purify total cytoplasm RNA.

2.7 Antioxidant enzyme assays

The activities of superoxide dismutases (SODs) and glutathione peroxidase (GPx) were evaluated as described previously [12]. The SOD activity assay was based on the method developed by McCord and Fridovich [13], modified by Flohé and Ötting [14]. GPx activity was measured according to the method of Lawrence and Burk [15] modified by Flohé and Günzler [16]. Cardiac protein levels were determined according to the method of Lowry et al. [17]. Enzyme activities were expressed in international units per mg of protein (I.U./mg Pt).

2.8 Non-enzymatic antioxidant assays

Total ascorbate concentrations were measured in plasma and heart homogenates by high-performance liquid chromatography using fluorimetric detection at 360 nm excitation and 440 nm emission, as previously described [18], and were expressed in μM in the plasma and nmol/g of tissue in the heart.

Vitamin E (α-tocopherol) was determined in plasma and hearts by a method derived from Burton et al. [19]. Vitamin E concentrations were expressed in μM of plasma and μg/g of tissue.

Uric acid concentration was determined in plasma using the Sigma enzymatic procedure (Sigma) using uricase and peroxidase and was expressed in μM.

2.9 Determination of oxygen radical absorbance capacity

Oxygen radical absorbance capacity (ORAC) was determined on the plasma according to a modified method of Cao et al. [20] and has been described in detail elsewhere [21]. Briefly, the reaction mixture contained a final concentration of 16.7·10−8 M β-allophycocyanin in 75 mM phosphate buffer, pH 7.0, at 37°C in the presence or the absence of Trolox or diluted plasma (dilution 1:500). The reaction was initiated by the introduction of 3·10−3 M 2,2′-azobis(2-amidinopropane)-4-hydrochloride, and followed spectrophotometrically by the decrease in fluorescence at 598 nm excitation and 615 nm emission. Trolox was used as a reference antioxidant for calculating the ORAC values, with one ORAC unit defined as the net protection area provided by 1 μM final concentration of Trolox.

2.10 Determination of lipid peroxidation in tissue and plasma

Lipid peroxidation determination was carried out on cardiac tissue and plasma by the thiobarbituric acid reactive species (TBARs) assay according to a modified method of Buege and Aust [22]. Cardiac homogenates and plasma were combined with the trichloroacetic acid–thiobarbituric hydrochloric acid reagent (13.5% w/v, 0.33% w/v, 0.85 M, respectively) and thoroughly mixed. The tubes were boiled for 15 min and cooled in a bucket of ice. Trichloroacetic acid (70%) was then added and after a 20 min incubation, the flocculent precipitate was removed by centrifugation. The fluorescent measurement was made at 515 nm excitation and 553 nm emission. TBARs were expressed as micromoles of malondialdehyde (MDA) per litre of plasma (μM) and in nmol/mg of protein in the heart (nmol/mg Pt).

View this table:
Table 1

Primer sequences and PCR conditions for β-actin, iHSP70, MnSOD, CuZnSOD and GPxa

 SequenceNucleotide localisationAmount of primers (ng)TA (°C)Number of PCR cycles
  • a F, forward; R, reverse; TA, annealing temperature.

2.11 Quantification of gene transcripts by RT-PCR

The steady state level of the studied gene transcripts has been evaluated by a comparative RT-PCR assay using the β-actin sequence internal standard as previously described [23]. Briefly, total cytoplasm RNA was purified from mechanically homogenised myocardium tissue using the Tri-Reagent (Sigma). The first-strand cDNA was synthesised by random priming from one microgram of total RNA using the MMLV reverse transcriptase (Gibco BRL, France). For reliable quantification, the concentrations of each set of primers and the stringency of PCR conditions were adjusted so that the amplicons would have similar amplification kinetics and their exponential phases of the amplification overlap [23]. Oligonucleotide primers were chosen with a similar melting temperature (Tm) and GC content using the PcrBase program (Bisance, Infobiogen) from Genbank extracted sequences (Table 1). Total RNA samples were pretreated with DNase I.

After amplification PCR products were electrophoresed in a 2% ethidium bromide stained agarose gel for quantification of fluorescence by image analysis under UV light. The PCR amplification yield of target sequences was expressed in arbitrary units (AU) as a ratio of the target gene to the β-actin electrophoretic bands optical density.

2.12 Statistical analysis

All data are presented as means±S.E.M. Tests of significance yielding statistical comparisons were performed with two factors fully factorial ANOVA, the two factors being the type of rat (ODS versus Wistar) and the presence of PBN. ANOVA was followed by inter group pair wise comparisons with Tukey HSD multiple comparisons. For biochemical, molecular and enzymatic measurements, for which only two groups are considered, statistical analysis was performed with a t-test, determining differences between Wistar and ODS rats.

View this table:
Table 2

Biochemical measurements on Wistar and ODS rat plasma; results are expressed as means±S.E.M.

 Wistar ratsODS ratsStatistical significance
Vitamin C (μM)26.24±1.921.43±0.09P<0.001
Vitamin E (μM)19.98±1.4515.42±0.58P<0.05
MDA (μM)6.28±0.273.44±0.09P<0.01
ORAC value2.49±0.102.37±0.11N.S
Uric acid (μM)12.49±2.9529.63±4.34P<0.01

3 Results

3.1 Rats

After 15 days of an ascorbic-acid-free diet, ODS rats showed characteristic features of scurvy: haemorrhage around eyes and nose, muscle and leg joint haemorrhages, difficulties in walking and poor general condition. Body weight and heart weight of ODS rats were significantly lower than Wistar rats (409.1±12.9 vs. 572.5±15.5 g and 1.27±0.02 vs. 1.39±0.02 g, respectively, P<0.01). Heart-to-body ratio was significantly higher for ODS rats (0.31%±0.01 vs. 0.24%±0.01, P<0.05).

3.2 Plasma biochemical measurements

Plasma biochemical measurements (Table 2) have shown that ascorbic acid concentration was markedly decreased (95%) in ODS rats compared with Wistar rats 2 weeks after a vitamin-C-deficient diet. Vitamin E plasma concentration was also reduced by about 20% in ODS rats (19.98±1.45 vs. 15.42±0.58 μM). However, MDA content was lower in the plasma of ODS rats (6.28±0.27 vs. 3.44±0.09 μM) and oxygen radical absorbance capacity was identical for both. Uric acid was shown to be significantly increased in ODS rat plasma (12.49±2.95 vs. 29.63±4.34 μM).

View this table:
Table 3

Functional parameters of Wistar and ODS rat hearts during preischemic normoxic perfusion. Measures of the functional parameters correspond to the values observed at the end of preischemic perfusion, for control hearts and hearts infused with 3 mM PBN; results are expressed as means±S.E.M.

  Heart weight (g)PCoronory flow (ml/min/g of heart)PLeft systolic ventricular pressure (mmHg/g of heart)P
Wistar 1.39±0.040.0114.39±0.54NS83.69±4.990.001
ODS 1.30±0.030.0113.97±0.93NS117.13±3.150.001
  Left ventricular developed pressure (mmHg/g of heart)PHeart rate (beats/min)PRate pressure product (103)P
Wistar 76.51±5.750.001261.67±12.760.0127.37±1.75NS
ODS 109.28±3.410.001231.67±14.700.0131.55±2.43NS

3.3 Experimental myocardial ischemia–reperfusion

3.3.1 Functional parameters

During preischemic normoxic perfusion, the hearts of vitamin-C-deficient ODS and Wistar rats demonstrated distinct evolution of functional parameters (Table 3). Although coronary flow was not different between the groups, left systolic ventricular pressure and left ventricular developed pressure (LVDP) were approximately 35 mmHg higher in ODS rat hearts (P<0.001), and were not influenced by the administration of PBN. However, the heart rate was significantly lower for ODS rat hearts (P<0.01), therefore the rate pressure product (RPP) remained the same between the groups.

Global total ischemia was induced by the stop flow of the aortic perfusion, and was rapidly followed by the cessation of myocardial contractions. Reperfusion resumed coronary flow up to 40% of its initial value but no statistical differences were observed among the groups after 30 min of reperfusion (Wistar: 31±5%; Wistar+PBN: 31±6%; ODS: 35±3%; ODS+PBN: 43±8%).

During ischemia, intra-ventricular pressure gradually increased for 20 min reaching values close to 50 mmHg which are representative for the phenomenon of ischemic contracture (Fig. 2), then decreased slowly. Upon the relief from 30 min of global ischemia, left diastolic end ventricular pressure (LDEVP) increased rapidly and reached a maximum 2 min after the onset of reperfusion, then decreased slowly but remained at a high level during the whole reperfusion period. This feature is characteristic of postischemic contracture and was shown to be statistically reduced in ODS rat hearts (P<0.01 at the end of reperfusion) and not by PBN treatment.

Fig. 2

Evolution of left diastolic end ventricular pressure (LDEVP) during ischemia–reperfusion protocol (values are related to heart weight and expressed in mmHg/g of heart). Results are presented as means±S.E.M. ○, Wistar rats (n=6); □, Wistar rats+PBN (n=6); ●, ODS rats (n=6); ■, ODS rats+PBN (n=5) significantly different from Wistar rat: *, P<0.05; **, P<0.01.

During ischemia, left ventricular developed pressure (LVDP) was reduced to zero (Fig. 3). With reperfusion, LVDP of Wistar rat hearts was only slightly restored and remained at a low level at the end of the reperfusion period. Conversely, LVDP of ODS rats hearts recovered rapidly and reached 75% of its preischemic value at the end of reperfusion. No differences were observed due to the infusion of 3 mM PBN.

Fig. 3

Evolution of left ventricular developed pressure (LVDP=LsystolicVP−LDEVP) during ischemia–reperfusion protocol (values are related to heart weight and expressed in mmHg/g of heart). Results are presented as means±S.E.M. ○, Wistar rats (n=6); □, Wistar rats+PBN (n=6); ●, ODS rats (n=6); ■, ODS rats+PBN (n=5) significantly different from Wistar rat: *, P<0.05; **, P<0.01; ***, P<0.001.

Fig. 4

Evolution of rate pressure product (RPP=LVDP·HR) during ischemia–reperfusion protocol (values are related to heart weight and expressed in mmHgbeats/min). Results are presented as means±S.E.M. ○, Wistar rats (n=6); □, Wistar rats+PBN (n=6); ●, ODS rats (n=6); ■. ODS rats+PBN (n=5) significantly different from Wistar rat: ***, P<0.001.

The evolution of the RPP during ischemia and reperfusion was found to show the same pattern as described for LVDP (Fig. 4). Hearts from ODS rat showed significantly higher recovery of LVDP than the control Wistar rat hearts (P<0.001) during reperfusion. When expressed as a percentage of the preischemic values (Fig. 5) the recovery of RPP at the end of the reperfusion period was still better for ODS rat hearts than for Wistar (40.4±7.2 vs.7.6±6.8, P<0.001).

Fig. 5

Evolution of the percentage of initial rate pressure product during the ischemia–reperfusion protocol. Results are expressed as means±S.E.M. ○, Wistar rats (n=6); □, Wistar rats+PBN (n=6); ●, ODS rats (n=6); ■, ODS rats+PBN (n=5) significantly different from Wistar rat: *, P<0.05; **, P<0.01; ***, P<0.001.

Fig. 6

Duration of contractile disturbances associated with reperfusion arrhythmias observed during 30 min of reperfusion. Results are presented as means±S.E.M. significantly different from Wistar rat: *, P<0.05; **, P<0.01.

Although correct classification of rhythm abnormalities must be defined according to the Lambeth's conventions on electrocardiographic criteria, monitoring ventricular pressure allowed us to evaluate the contraction abnormalities that reflect reperfusion arrhythmias. Ventricular contractile disturbances are frequently observed after 30 min of a global normothermic ischemia with an average duration of 20.8±5.3 min in Wistar rat hearts (Fig. 6), and are mostly represented by tachycardia and fibrillation. The duration of these disturbances was significantly reduced in ODS rat hearts compared with Wistar (P<0.01), and was not influenced by PBN treatment.

Fig. 7

Free radical release during reperfusion. (A) Spin adduct concentration in coronary effluents during preischemic normoxic perfusion and during reperfusion for Wistar (□, n=6) or ODS (■, n=5) rat hearts. Results are presented in nM as means±S.E.M. (B) Total cumulated spin adduct release during 30 min of reperfusion in nmol/min/g of heart. Results are presented as means±S.E.M.

3.3.2 ESR spin trapping

Experiments performed with ESR on coronary effluents showed the presence of a sextuplet signal (aN=13.5 G, aH=2.1 G, g=2.012) with coupling constants that could be attributed to alkyl/alkoxyl spin adducts. The concentration of these free radical species in the coronary effluent was evaluated (Fig. 7A). During normoxic preischemic perfusion a low spin adduct was observed in coronary effluents, however, after reperfusion, a large release of alkyl/alkoxyl species occurred and remained at a high level during 30 min of reperfusion. No differences in terms of spin adduct concentration or cumulated spin adduct release (Fig. 6B) were observed between ODS and Wistar rat hearts.

3.4 Biochemical measurements in heart tissue

Biochemical measurements (Table 4) in heart tissue after the ischemia reperfusion event have demonstrated that ascorbate content of ODS rats corresponded only to 5% of its value in Wistar heart (591±54.9 vs. 31.9±3.3 nmol/g). However, vitamin E and MDA tissue contents were not different between ODS and Wistar rats.

3.5 Gene expression

In order to examine gene expression, the relative mRNA abundance of antioxidant enzymes and of the inducible form of heat shock protein 70 (iHSP70) were assessed on rat hearts taken at the end of ischemia–reperfusion protocol. The genes of the enzymes responsible for superoxide radical anion degradation in mitochondria and cytoplasm, respectively MnSOD and CuZnSOD, were expressed at a higher level in ODS rat hearts (Fig. 8A and B). Moreover, the mRNA expression of GPx (Fig. 8C) and iHSP70 (Fig. 8D) was statistically greater in ODS compared with Wistar rat hearts.

View this table:
Table 4

Biochemical measurements and antioxidant enzymes activities of Wistar and ODS rat heart tissue untreated with PBN; results are expressed as means±S.E.M.

 Wistar ratsODS ratsStatistical significance
Vitamin C (nmol/g)591.0±54.931.9±3.3P<0.001
Vitamin E (μg/g)55.23±5.3060.30±4.47NS
MDA (nmol/mg Pt)3.56±0.274.74±0.40NS
MnSOD (I.U./mg Pt)83.68±22.5153.29±10.20NS
CuZnSOD (I.U./mg Pt)12.97±3.2511.41±3.14NS
GPx (I.U./mg Pt)1.14±0.191.02±0.11NS

3.6 Enzymatic activities in heart tissue

Myocardial antioxidant enzyme activities in Wistar and ODS rat hearts were processed after the complete ischemia–reperfusion period (Table 4). No differences in SODs and GPx activities were observed between ODS and Wistar rat hearts.

Fig. 8

mRNA expression of antioxidant enzymes and iHSP70 for Wistar (n=6) or ODS (n=6) rat hearts. Results are presented as means±S.E.M. (A) CuZnSOD; (B) MnSOD, (C) GPx; (D) iHSP70.

4 Discussion

The aim of this study was to determine the interaction of tissue ascorbic acid levels with the oxidative stress associated with a myocardial ischemia–reperfusion event. Hearts of ODS rats were used because this mutant strain of Wistar rats lack the key enzyme responsible for hepatic synthesis of ascorbic acid, and they can become scorbutic with a vitamin-C-deficient diet. ODS ascorbic acid deficient rat hearts were compared with healthy Wistar rat hearts as this latter is widely used for in vitro and in vivo experimental studies in the field of myocardial ischemia–reperfusion injury.

Two weeks after a vitamin C free diet, ODS rats showed characteristic features of scurvy syndrome with haemorrhages, poor general condition and decrease in body weight. Indeed, we observed a dramatic decrease in ascorbate concentrations in plasma and in heart tissue of ODS rats, that was consistent with previous reports [24]. Paradoxically, this general decrease in vitamin C was not accompanied by increased markers of lipid peroxidation (MDA levels) in plasma and myocardial tissue. Tanaka et al. [24] reported that an increase of TBARs in heart and plasma of ODS rats can only be observed after 21 days deficiency in vitamin C, at which time vitamin E levels are also markedly impaired. These observations are consistent with the results observed in our study. Moreover, despite the lack of vitamin C, the antioxidant capacity of ODS plasma was maintained at the level observed in Wistar rats. Wayner et al. [25] have shown that ascorbate was not the major component contributing to the antioxidant capacity of human blood plasma, and that uric acid was very potent in scavenging free radicals. Uric acid might contribute up to 60% of the measured total antioxidant potential of plasma in healthy subjects and is reported to bind iron [26]. Additional experiments done in our laboratory on Wistar rat plasma in which ascorbate concentration was decreased to the levels observed in ODS rats, through ascorbate oxidase treatment, showed that the ORAC value was impaired and could be restored by the addition of ascorbic acid (data not shown). We have demonstrated here that uric acid levels were increased in ODS rats, which may therefore explain why, despite the decrease in antioxidant vitamins (major for vitamin C and minor for vitamin E) the antioxidant capacity of plasma was not altered and lipid peroxidation was not increased in these rats. These results also showed that, in the blood of vitamin-C-deficient rats, other alternative antioxidant defences might be induced to counterbalance the loss. In vivo, the relative contribution of each of these antioxidant agents, ascorbate and uric acid, to the total antioxidant capacity of a tissue is not fully understood. Furthermore, these antioxidants are believed to act cooperatively in vivo, so as to provide greater protection against free radical mediated injury than would be afforded by any single antioxidant alone. The ways by which uric acid is increased in ODS vitamin-C-deficient rat plasma are currently under investigation.

Ascorbate is regarded as one of the most potent water-soluble antioxidants present in cells and body fluids [7], efficient in scavenging free radicals produced in different conditions of oxidative stress. Therefore, a loss of vitamin C tissue availability might impair its ability to counterbalance a burst of free radical production, as observed during postischemic reperfusion in the heart [6,27] and that administration of exogenous antioxidant compounds would improve the recovery [28]. During preischemic normoxic conditions, ODS rat hearts were shown to have a distinct contractile function from Wistar rat hearts, with a higher LVDP and lower heart rate, but not a different RPP. After 30 min of global total ischemia, ODS rat hearts recovered quickly for the ischemic insult, with lower LDEVP, higher recovery of LVDP and RPP and less occurrence of rhythm disturbances. These results provide evidence that vitamin C-deficient ODS rat hearts have a different mechanical function and contractile responsiveness to reperfusion injury and demonstrate for the first time that ascorbic acid deficient rat hearts have a better recovery of their myocardial function during postischemic reperfusion. Moreover, in ascorbate-deficient hearts the functional modifications associated with reperfusion were not accompanied in our study by increased levels of free radical release. Our findings are therefore in contradiction with the common statement that maintaining myocardial hydrophilic antioxidants may aid in the protection of the myocardium [6]. Alternatives pathways of protection might compensate for the lack of vitamin C and it is possible that one antioxidant may equilibrate with an other, and that all endogenous antioxidant defences may act in concern to protect the tissue against oxidative injury.

Ascorbyl free radical, the radical derivative of vitamin C, can be considered as an indicator of oxidative stress [29,30]. In previous studies we had shown that ascorbyl free radical could be detected in the coronary effluent of isolated rat hearts during postischemic reperfusion and that the amount of ascorbyl free radical released was closely related to the impairment of myocardial function [4]. Therefore, a vitamin C-deficient tissue might be more prone to oxidative stress and would release larger amounts of free radicals during a postischemic reperfusion. Paradoxically, we have found here that ODS rat hearts do not release higher amounts of free radical species than those of Wistar rats, and that their MDA tissue content was not different, showing that the intensity of lipid peroxidation was not different between the two groups. These surprising results raise again the question of the exact in vivo role of vitamin C as a potent free radical scavenging component. In other studies processed in vitro, ascorbate was shown to exert pro-oxidant effects, especially in the presence of transition metals which are reduced back to their active catalytic form by ascorbate during oxidative reactions [31,32]. For instance, the association of ascorbate and iron or copper is a useful way to induce general oxidative stress in animals [33] or to generate hydroxyl radicals in buffers [34]. During myocardial ischemia reperfusion, it was shown that iron and copper are released from the tissue as free catalytic metals [35]. Therefore, in ODS rat hearts, the lack of tissue ascorbate during reperfusion might be responsible for a reduction of oxidative stress and lipid peroxidation chain reactions triggered by free iron and copper present in and between the cells.

Despite their low vitamin C tissue concentration, ODS rat hearts which had been submitted to 30 min ischemia and 30 min of reperfusion were shown to have a larger expression of antioxidant enzymes like CuZnSOD, MnSOD and GPx than Wistar rat hearts. However, this up-regulation of gene expression was not accompanied by a corresponding increase in tissue enzyme activities. This uncoupling has been recently demonstrated in cells over expressing the transfected MnSOD gene [36] and, in our laboratory, in rat heart ventricles after myocardial infarction [23]. The reduction of myocardial postischemic dysfunction observed in ODS vitamin C-deficient rat hearts was associated with higher expression of iHSP70 than in Wistar rats. Our results indicate that genes contributing to myocardial protection against the deleterious effects of ischemia and reperfusion are likely to be implicated in the cardioprotection observed in ODS rat hearts. Heat shock proteins have been extensively shown to provide tolerance against oxidative stress and protection against the ischemic insult. Bornman et al. [37] have recently suggested that the preferential target for HSP70-mediated protection against oxidative stress was the preservation of state-3 respiration in the mitochondria. Several studies have shown that initial stressful events, also named preconditioning, led to the enhancement of endogenous defence systems via up-regulation of heat shock or antioxidant enzymes expression, rending the heart more tolerant to ischemia–reperfusion injury [38]. Myocardial preconditioning can be triggered by different ways, ischemia, heat, pharmacological treatment but also free radicals [39,40]. During the initial phase of ascorbic acid deprivation in ODS vitamin C-dependent rats, the lack of this antioxidant vitamin could be responsible for a general increase of oxidative stress in blood and tissues, which could in turn trigger the activation of compensatory mechanisms related to the general term of preconditioning.

Different authors have observed that ODS rats fed with an ascorbic acid deficient diet have lower activities of hepatic-drug metabolising enzymes [10], and higher levels of serum cholesterol [41] and corticosterol [10] than Wistar rats. The profound impact of corticosteroids on nuclear transcription and protein synthesis is well recognised, and receptors for glucocorticoids exist in vascular smooth muscle and endothelial cells [42,43]. Moreover, the anti-inflammatory properties of glucocorticoids could account for the lesser response to oxidative stress noted in ODS rat hearts. Therefore, the protective effect of chronic high blood corticosterol levels cannot be excluded in the ODS model, and is still under investigation. However, it is proposed that some of the endogenous antioxidants act as a primary defence mechanism whereas the others including vitamins and plasma compounds play a secondary role for reducing the ischemia–reperfusion injury [44].

In conclusion, our study has shown for the first time that vitamin-C-deficient ODS rat hearts are more resistant than healthy Wistar to an ischemia–reperfusion insult, with higher recovery of contractile function and similar levels of free radical release during the reperfusion period. These experiments show that ascorbic acid deficient ODS rats are probably developing alternative antioxidative defences against the oxidative stress related to ischemia and reperfusion, and raises the question of the relative importance of each endogenous antioxidant in the resistance of the heart to ischemia–reperfusion injury.


A Young Investigator Award has been attributed to this study by the Oxygen Society annual meeting 1999. The authors gratefully acknowledge the Conseil Régional de Bourgogne for its continuing support. The authors thank Dr. Paul Walker for correcting the English in the manuscript.


  1. [1]
  2. [2]
  3. [3]
  4. [4]
  5. [5]
  6. [6]
  7. [7]
  8. [8]
  9. [9]
  10. [10]
  11. [11]
  12. [12]
  13. [13]
  14. [14]
  15. [15]
  16. [16]
  17. [17]
  18. [18]
  19. [19]
  20. [20]
  21. [21]
  22. [22]
  23. [23]
  24. [24]
  25. [25]
  26. [26]
  27. [27]
  28. [28]
  29. [29]
  30. [30]
  31. [31]
  32. [32]
  33. [33]
  34. [34]
  35. [35]
  36. [36]
  37. [37]
  38. [38]
  39. [39]
  40. [40]
  41. [41]
  42. [42]
  43. [43]
  44. [44]
View Abstract