Table 1 shows characteristics of rats during 7 stages of development. During perinatal development, as expected, there was a progressive increase of heart weight (HW) and body weight (BW). The heart weight/body weight ratio decreased significantly from 10.9 ± 2.1 mg heart/g body weight in fetuses to 3.94 ± 0.1 mg heart/g body weight in adults (p < 0.001). Heart rate (HR), obtained in perfused heart experiments (13 days old onwards) was unchanged over time. Left ventricular developed pressure (LVDP) was not significantly different in isolated, buffer-perfused hearts.

The mean plasma creatine concentrations, which could be obtained from 7 day old animals onwards, are also shown in Table 1. During maturation there was a steady decrease of up to 44% in plasma creatine concentration (0.97 ± 0.07 mmol/L (7 days) vs 0.54 ± 0.08 mmol/L (80 days), reaching significance (p < 0.05) from 33 days of age onwards.

Total creatine content (Figure 1A) in the heart significantly (ANOVA p < 0.001) increased during maturation from 30.1 ± 1.3 nmol/mg protein in -2 day old rats to 87 ± 2.9 nmol/mg protein in 80 day old rats, corresponding to a ~3-fold increase. Similarly, total creatine kinase (CK) activity increased several-fold (p < 0.001) during maturation (2.1 ± 0.26 mIU/mg protein (-2 days) vs 8.06 ± 0.4 mIU/mg protein (80 days) (Figure 2)). Compared with adult heart, B-containing CK isoenzymes are relatively more abundant in foetal myocardium, but total CK activity is much lower. Calculating CK isoform activities from total CK and % isoenzyme distribution revealed that the BB isoform did not increase until 80 days of age (ANOVA: p = 0.01), whilst MB- (2.1-fold, p < 0.001) and MM-isoforms (4.7-fold, p < 0.001) increased directly after birth. The activity of mito-CK, which was not detectable until at 13 days after birth, increased during growth from 0.0075 ± 0.05 to 2.68 ± 0.26 mIU/mg protein. Citrate synthase activity, a marker of mitochondrial volume, rose significantly (p < 0.001) in a steady manner throughout the perinatal period and reached maximum levels in the adult of 0.95 ± 0.03 mIU/mg protein.

Figure 3 shows the developmental pattern of total lactate dehydrogenase (LDH) activity. Although no significant difference was found in total LDH activity during maturation, the distribution of LDH isoenzyme activities changed significantly. LDH₁, LDH₂ and LDH₃ isoenzymes showed a significant increase (~7.4, ~3.6 and 1.4 fold respectively), while LDH₄ and LDH₅ isoenzymes decreased (~3 and 12.5 fold respectively). This shift in LDH isoenzyme activity reflects the change from anaerobic to aerobic metabolism during maturation.

To explore changes in CrT mRNA expression during maturation, the total RNA isolated from rat hearts of different age groups was analyzed using a real-time quantitative RT-PCR technique. The abundance of the transcripts for CrT increased steadily with age (Figure 1B). At 80 days of age, cardiac mRNA levels of CrT were significantly higher compared with all other experimental groups, showing a ~3.6 fold higher relative expression of CrT as compared to -2 day old rats (ANOVA p < 0.001).

CrT function was examined by measuring creatine uptake by means of ¹⁴C creatine labeled heart perfusion. Table 1 shows the rate of creatine uptake at 500 μmol/L extracellular creatine in 13 to 80 day old rats. Creatine uptake steadily increased with age and was significantly higher than in 13 day old hearts from day 50 onwards. Creatine uptake was 3.0 ± 0.6 nmol/min/gww in the 13 day old heart and reached 4.9 ± 0.5 nmol/min/gww at 80 days of age, i.e. increased by ~65%. Thus, creatine content, CrT expression and CrT function all increased steadily during maturation.

There was a significant correlation (r² = 0.70, p < 0.001) between CrT mRNA and total creatine levels (Figure 4A) as well as CrT mRNA and creatine uptake (r² = 0.90, p < 0.001, Figure 4B).

The velocity of the CK reaction changes with the size of the creatine pool and also with the accumulation of CK isoenzymes and the shift in isoenzyme distribution 14 . The appearance of CK function in myofibrils and mitochondria during maturation has been studied in a number of animal models 16 12 9 10 . In agreement with previous studies 13 14 , we showed that mito-CK, bound on the inner membrane of the mitochondria, and MM-CK, a significant portion localised at sites of ATP utilisation, e.g. myofibrils, are the predominant isoenzymes in mature myocardium, whereas BB-CK and MB-CK isoenzymes dominate in the immature myocardium. Adult levels of mito-CK are not reached until late in development (50 days), suggesting that its role in mediating PCr synthesis from oxidative phosphorylation is required at a relatively late stage of maturation. These findings confirm previous results and show that the maturational changes in creatine metabolism described occur in parallel to the expected changes in the creatine kinase system.

In line with previous studies 17 , total creatine content in hearts increased 3-fold during maturation. Recent findings from Ireland et al 19 in the spiny mouse, which, unlike conventional rats and mice, has a long gestation of 38-40 days, revealed that the estimated amount of total creatine did not change in the foetal heart between 30-37 days gestation, but it increased significantly by postnatal day 10. In the fetus, myocardial creatine is derived both from the mother and from de novo synthesis. During maturation myocardial creatine levels increase as the demand for all energy-requiring processes in the heart rises. Likewise, myofibrillar ATPase activity increases steadily during the perinatal period 12 . Thus, energy producing and energy utilising reactions are concomitantly increased within the cardiomyocyte.

In our study, changes in total LDH activity were accompanied by distinct changes in LDH isoenzyme pattern. Specifically, we observed an increase in LDH₁, LDH₂ and LDH₃ isoenzymes. LDH₁ and LDH₂ were predominant in the adult heart, whilst LDH₄ and LDH₅ were found to dominate in the foetal heart. These changes are again in line with previous findings 21 and are indicative of the shift from anaerobic to aerobic metabolism during maturation.

Similarly, citrate synthase, a marker of mitochondrial volume, increased during maturation. A rise in mitochondrial numbers is consistent with previous studies 22 and reflects the increased energy demand of mature myocardium.

Studies using cultured cells 23 24 25 , isolated rat hearts 26 and in vivo human studies 27 indicate that muscle cells tightly regulate their intracellular creatine content. However, the underlying cellular and molecular mechanisms by which these cells perform this task are only partially understood. The two enzymes required for creatine biosynthesis L-arginine:glycine amidinotransferase and S-adenosylmethionine:N-guanidinoacetate methyltransferase 28 may be expressed in heart tissue 29 , but activities of these enzymes are negligible and therefore creatine is not synthetised in heart muscle in significant amounts. Instead, creatine is taken up into the heart by the action of the creatine transporter. During maturation, cardiac creatine levels increase steadily. Therefore, we hypothesized that during development the CrT-governed creatine uptake is progressively increased.

Indeed, we found the maximal rate of creatine transport to steadily increase during maturation, in parallel with both total creatine content and creatine transporter mRNA levels. Importantly, we found a significant correlation (r² = 0.70) between CrT mRNA and total creatine levels as well as CrT mRNA and creatine uptake (r² = 0.90). Thus during development, 90% of the total variance in creatine uptake capacity and 70% of the variance in creatine content could be explained by changes in gene expression of the CrT. This is in contrast to the adult heart, where changes in CrT activity in response to genetically altered creatine content were mainly driven by non-transcriptional means, e.g. substrate feedback or post-translational modification 30 .

The near maximum rate for ¹⁴C-creatine uptake in isolated perfused 80 day old rat hearts (4.99 ± 0.52 nmol/min/gww) agreed with previously published data on adult rat heart 31 26 32 . The observed increase in heart weight does not explain the observed maturation-induced increase in Na⁺/Cl^- creatine transport activity because the uptake values were normalised to mg of protein. Likewise, Garcia-Delgado et al 33 found an up-regulation of renal Na⁺/Cl^- creatine transporter during ontogeny in rats. In their study, age increased V_max, indicating that renal maturation may have increased the density of creatine transporters and/or their turnover rate.

Similar to our findings, Braissant et al 34 showed detectable CrT mRNA levels in the rat embryo at 18.5 day of gestation, while at day 12.5 and 15.5, mRNA levels were absent. Ireland et al 19 recently demonstrated in the spiny mouse that, between gestational day 37 and postnatal day 10, cardiac mass increased almost 3-fold and total creatine content increased 2-fold. CrT mRNA expression, however, did not increase significantly between mid-gestation and gestational day 37 and stayed at similar levels until postnatal day 10. Thus, the authors suggest that creatine uptake into the heart is facilitated by an increase in CrT activity rather than CrT protein expression. In contrast, we found a significant increase of CrT mRNA expression in the rat heart at day 13 compared to pre-gestation (-2 days).

The reason for decreasing plasma concentration of Cr at different ages may be related to the fact that Cr accumulates in several organs and tissues (e.g. heart, brain 35 and skeletal muscle 36 at later stages of development, likely trough an increase in Cr uptake. Higher uptake may in turn lead to lower Cr levels in the plasma as observed in the present study. In fact, for the heart we have found an inverse relationship (r² = 0.588, p < 0.05) between plasma Cr and CrT gene expression. A decrease in intestinal Cr uptake during development 37 may add to this phenomenon of decreasing plasma concentration.

Furthermore, phosphorylation of CrT by cAMP-dependent protein kinase A and protein kinase C, which affect CrT activity, may be an underlying signal transduction pathway by which the CrT in heart is regulated during early development after birth as suggested by others 37 .

Unfortunately, to date it remains impossible to determine how the increases in CrT mRNA, activity and creatine content are paralleled by an increase in CrT protein content, since all available anti-CrT antibodies cross-react with pyruvate dehydrogenase and other proteins 20 . Therefore, determining the changes in CrT protein expression during development including post-translational regulatory mechanisms remains an important milestone for the future. Another limitation of our study is that we did not perform measurements of high-energy phosphate metabolites ATP, phosphocreatine, and of the free energy change of ATP hydrolysis (delta-G). However, on wet chemical analysis, phosphocreatine is highly unstable and it would be a challenge to measure this reliably in the smallest hearts. Furthermore, delta-G requires measurement of intracellular pH by ³¹P-MRS in isolated perfused hearts, which is currently not technically possible in the youngest age groups. These analyses are important goals for future research.

All experimental procedures in this study were approved by the University of Oxford Animal Ethics Review Committees and by the Home Office (London, U.K). Rats used in all studies were of the Wistar strain. Timed pregnant rats were obtained from a commercial breeder (Harlan, U.K) and kept under controlled conditions for temperature, humidity and light, with creatine-free rat chow and water available ad libitum. Pups were weaned at 21 days and placed on the same chow as the mother. Fetuses at 18 days after gestation (-2 days) were obtained by Caesarean section after anesthetising the mother with 1.5-2% isoflurane. Animals of age 7, 13, 21, 33, 50 and 80 days were used.

Animals were anesthetized with an intraperitoneal injection of 75 mg/kg body wt sodium pentobarbitone (Sagatal, Rhône Mérieux, Dublin), whilst foetuses were sacrificed by cervical dislocation. After loss of pedal reflexes, blood samples were removed from the femoral vein (animals aged 7 days onwards) and hearts were rapidly excised and arrested in ice-cold Krebs-Henseleit buffer (see below). Hearts from 13, 21, 33, 50 and 80 day old rats were cannulated via the ascending aorta for retrograde perfusion radiolabelled ¹⁴C-creatine transport studies and metabolite assays (-2 and 7 day old hearts were too small for perfusion). A water-filled cling film balloon, attached via polytetrafluorethylene tubing to a Gould disposable pressure transducer, was inserted into the left ventricular cavity via the mitral valve, secured by a ligature and inflated sufficiently to give an end-diastolic pressure (EDP) of ~4 mm Hg. Heart rate and left ventricular (LV) pressures were recorded continuously using a MacLab 4a (AD Instruments Ltd, Hastings, UK) linked to Chart™ software (AD Instruments).

Hearts from > 13 days old rats (different from the rats used for the biochemical analysis, which in our setup is not possible on radioactive samples) were perfused with Krebs-Henseleit (KH) buffer containing (mmol/L) NaCl (110), KCl (4.7), MgSO₄.7H₂O (1.2), CaCl₂.2H₂O (1.75), glucose (11), KH₂PO₄ (2), NaHCO₃ (25), Na₂EDTA (0.5), Na-Lactate (0.5) and Na-Pyruvate (4.5). The buffer, supplemented with 500 μmol/L creatine, was gassed with 95% O₂/5% CO₂ to give a pH of 7.4 at 37°C. Creatine concentration was used at the optimal level of 500 μmol/L, since Cr uptake by adult rat hearts at this concentration, approaches Vmax 26 31 . Hearts were perfused with a recirculating volume of ca. 400 ml at a constant pressure of 100 mmHg. Cardiac contractile function was monitored throughout the protocol as described above. Since the accumulation of ¹⁴C-labelled creatine in the heart has been shown to be linear throughout a 1 hr incubation 26 , 30 minutes was used as the optimal radioactive label perfusion time. Hearts were allowed to equilibrate for 10 min and then ¹⁴C-labelled creatine (Apin Chemicals Ltd, Abingdon, U.K) was added in trace amounts to the recirculating buffer. Hearts were perfused in the presence of ¹⁴C-creatine for 30 minutes, after which time a final 15 min label washout period followed, using a non-circulating mode with non-radioactive buffer. The absence of ¹⁴C-label in the effluent between 10 and 15 min of washout ensured that the extracellular space had been cleared and that creatine leakage from the heart was negligible. At the end of the perfusion protocol, hearts were weighed and scar tissue removed. Subsequently, hearts were digested in 1 ml of 1 M KOH at 60°C before scintillation counting for [¹⁴C] as described previously 26 .

Total RNA was extracted from frozen heart tissue using the RNeasy Kit (Qiagen), including treatment with proteinase K according to manufacturer's instructions. Total tissue RNA dissolved in RNase-free water was used in real-time RT-PCR (Qiagen Quantitect SYBR Green RTPCR kit, Qiagen, Crawley, United Kingdom), along with the following primers: RT-CrT-F 5'-gccggcagcatcaatgtc-3', RT-CrT-R 5'-ggtgttgcagtagaagacgatcac-3'. Quantitative fluorescent real-time RT-PCR analysis was performed to investigate mRNA expression levels in rat hearts of different age using the Rotor-Gene system (Corbett Research Ltd., Cambridge, United Kingdom). To demonstrate linearity of the reactions, a standard curve was generated using a series of total RNA dilutions from 80-day-old rat hearts. Results were normalized to the expression levels of the housekeeping gene β-actin (RTactinF 5'-gacaggatgcagaaggagattact-3', RTactinR 5'-tgatccacatctgctggaaggt-3'). Real-time RT-PCR was performed in duplicate with 8-10 ng of total RNA as input. Quantification was carried out using appropriate software to generate standard curves, which expressed relative quantities of PCR products in the experimental samples in arbitrary units relative to the standard curve.

Total blood samples were centrifuged at 3,000 rpm for 10 min. The supernatant was used for the determination of plasma concentrations of creatine by HPLC as previously described 38 .

Hearts were perfused for 10 minutes before being rapidly freeze-clamped using Wollenberger tongs. Aliquots were taken to measure protein content by the method of Lowry 39 . 0.1% Triton X was added and enzyme activities of citrate synthase, CK and LDH were measured using an Ultraspec 2100 pro (Pharmacia Biosystems) as previously described 37 . CK isoenzyme distribution was measured using the Rapid Electrophoresis System (REP, Helena Biosciences, UK) as a separation unit, and the REP CK isoforms kit (Helena Biosciences, UK). For measurements of LDH isoenzymes, the Titan Gel LD Isoenzyme System (REP, Helena Biosciences, UK) was used. The Electrophoresis Data Center (EDC, Helena Biosciences, UK) automatically quantified the separated isoenzyme band. For HPLC analysis, tissue was homogenised in 0.4 N perchloric acid at 4°C, neutralised and centrifuged for 5 min at maximum speed, at 4°C. The supernatant was used to measure total creatine levels as previously described 40 . Tissue concentrations were expressed in nmol/mg of protein.

Data are expressed as the mean ± SE and sample size is given in brackets. To determine statistical significance between age groups, test for linear trend in one way analysis of variance (ANOVA) was performed, followed by a Bonferroni multiple comparison test. Every time point was compared with every other time point. In case of inhomogeneity of variance the Dunnet T3 test was used. P values less than 0.05 were considered significant. Correlation analysis was conducted for selected variables. All statistical analyses were performed by SPSS (Version 13.0).