Main

The incidence of clinical seizures is highest in the newborn period(1). At this developmental stage seizures have many causes, with hypoxia and ischemia thought to be the most common(2). In rat pups hypoxia produces seizures most frequently at 10-12 d of age with lower frequencies at 5 and 20 d(3). This age is generally a period of brain hyperexcitability(4). Rat brain cellular energy metabolism increases between about 5 and 25 d of age as indicated by increasing rates of tissue aerobic glycolysis(5,6). The maturational increase in brain energy metabolism is paralleled by brain PCr/NTP ratio measured in vivo by 31P NMR spectroscopy. This ratio increases between 5 and 25 d of age in the rat pup(7,8). This increase in PCr/NTP ratio is due primarily to increased PCr, because the brain NTP concentrations, including ATP, do not change over this age period(9). In the human brain, the PCr/NTP ratio increases between 24 wk postconception to 2-3 mo postterm(10). Brain PCr/NTP ratios are approximately the same in 10-12-d-old rats and human term newborns, the ages of high seizure susceptibility. During seizures PCr increases in mature, but not in immature, cerebral gray matter(11,12). These observations suggest that low Cr or PCr is important in susceptibility to hypoxic seizures in the metabolically immature brain. To test this hypothesis, rat pups were injected with Cr for 3 d before exposing them to hypoxia on postnatal d 10 or 20. Before and during hypoxia, the electrocortical (EEG) activity or31 P NMR spectra were measured.

METHODS

Animals. Long Evans rats were studied under hypoxic conditions at 10 or 20 d of age. Litters were obtained at 5 or 15 d (Charles River Breeding Laboratories, Wilmington, MA) and housed in individual cages in the Animal Care Facility at Children's Hospital. At 7 or 17 d, each litter was divided into two groups, half receiving Cr in saline (3 mg/g of body weight) and half receiving the same saline volume s.c. once daily for 3 consecutive days. The care and use of animals complied with institutional and federal guidelines.

EEGs. The EEG electrodes (Samtec, New Albany, IN) were implanted epidurally under ether anesthesia. Burr holes were placed 1.5 mm on each side of the sagittal suture midway between the bregma and lambdoid sutures. Electrode plugs were mounted on the skull using fast acting adhesive and dental acrylic. Subcutaneous needle electrodes were placed on either side of the dorsal trunk to record heart rate. All recordings were on a Grass model 8-16 Polygraph (Grass Instruments, Quincy, MA) after recovery from ether anesthesia. Rats were lightly restrained and placed on a heating pad(Seabrook Medical Systems, Cincinnati, OH). Body temperature was maintained between 32 and 35°C.

NMR. Spectra were acquired at 145.587 MHz using an Oxford 8.9-cm vertical bore superconducting magnet (8.45 T)(8). The 90° excitation pulse was 10 µs. Each spectrum was the sum of 56 acquisitions with either 10-s (fully relaxed) or 2-s (partially saturated) interpulse times. During and after hypoxia, partially saturated spectra were acquired every 2 min. Changes in ATP were estimated from the β-NTP peak area and PCr from its peak area as measured by Lorentzian curve fitting and peak integration (Felix 2.3, Biosym Technologies, San Diego, CA). Intracellular pH was calculated from the chemical shift of the inorganic phosphorus peak(13).

The CK (EC 2.7.3.2)-catalyzed reaction rates were measured using the steady state saturation transfer experiment(8,14,15). Transfer of phosphoryl groups from PCr to γ-ATP was measured by selective saturation of the γ-NTP signal using a 3-s low power pulse followed by the nonselective excitation pulse. A second spectrum was acquired with the saturating pulse an equal distance on the opposite side of PCr. Phosphorus fluxes in the complex CK-catalyzed reaction PCr- + MgADP- + H+ [harr ] Cr + MgATP+ were analyzed as a first order exchange between PCr and ATP(15). The pseudo-first order rate constant (k) for the exchange from PCr to ATP was calculated from the chemical flux(J), PCr signals in saturated (Ms) and control(Mo) spectra, and the 3-s longitudinal relaxation time for PCr (T1)(8). kf= J/[PCr] = 1/(T1)PCr(Mo -Ms)/MS

Experimental design. One group of rats (seven experimental and seven controls at each age) was studied by 31P NMR spectroscopy. A fully relaxed spectrum, a partially saturated spectrum, and a steady state saturation transfer study of the CK-catalyzed rate constant were acquired with the urethane-anesthetized (1250 mg/kg of body weight, i.p.) rat breathing room air. A 4% O2 gas mixture then was introduced, and four partially saturated spectra were acquired over 8 min. Room air was reintroduced, and partially saturated spectra were acquired for at least 20 min. A steady state saturation transfer experiment then was repeated.

In a second set of 10- and 20-d-old rats treated with Cr or saline, EEGs, heart rates, and body temperature were recorded before, during, and after breathing 4% O2 for 8 min. Two groups of rats were studied. An initial 10-d-old group received ether for electrode placement and then urethane for the recordings. The urethane anesthesia was given to parallel the conditions in the NMR experiment in which the death rate was high. A second group, including 10- and 20-d-old rats, received ether for placement of scalp electrodes and no urethane. This group was studied to avoid masking the electrographic seizures with anesthesia. After recovery from the ether, baseline recordings were acquired with the animal breathing room air. Then, N2 was infused into the chamber until the O2 concentration reached 4% as determined with an oximeter (Instrumental Laboratories, Lexington, MA). The O2 concentration was maintained at 4% for 8 min after which room air again was infused into the chamber. Recordings were continued for at least 20 min after the hypoxic period.

Statistics. Prehypoxia PCr/NTP ratios, pH values, and CK rate constants were compared between Cr-injected rats and controls at each age using a t test for unmatched samples. Changes in PCr and ATP values during hypoxia were compared with prehypoxia values using repeated ANOVA (SAS Institute, Cary, NC). The CK rate constants were compared before and after hypoxia in control and Cr-injected rat pups using a paired t test. Frequency of seizures and death rates in control and treated pups were compared using Fisher's exact test for 2 × 2 contingency tables.

RESULTS

Controls and Cr-injected pups showed similar activity, coat characteristics, and body weights at 10 and 20 d. Before hypoxia, no control or treated pups died. Heart rates (Table 1) and body temperatures (not shown) were the same in both groups. The prehypoxic brain PCr/NTP ratio was 25% higher in treated pups than in controls at 10 d, but was the same in the two groups at 20 d (Table 1). As expected, this ratio was higher at 20 than at 10 d in control animals(7,8). The mean ratio in 10-d-old Cr-treated pups approached the value found in the 20-d-old rats. Brain pH values were the same in control and treated pups at both ages. Rate constants for the CK catalyzed reaction in brain were much higher at 20 than at 10 d, a difference which was similar to that previously described(8). The rate constant was not different in control and treated pups at either age.

Table 1 The effects of creatine on heart rate and in-vivo31 P NMR measures of brain CK reactants and reaction rates in 10- and 20-d-old rats
Full size table

Pretreating rats with Cr significantly decreased hypoxic deaths at 10 but not 20 d (Table 2). When anesthetized with urethane for NMR studies, all but one of the 10-d-old controls died, whereas only one of seven treated rats died. All deaths occurred between 4 and 6 min after the hypoxic period. In the urethane-anesthetized 10-d-old pups studied outside the magnet, no gasping was seen in controls after 1-2 min of hypoxia. All Cr-treated pups showed ventilatory efforts throughout the hypoxic period. Heart rates in controls dropped to 60-100 beats/min, whereas heart rates in treated 10-d-old pups dropped only 20% to about 400 beats/min (p< 0.01).

Table 2 The effects of creatine injections on hypoxia-induced seizures and mortality in rats studied at 10 and 20 d of age
Full size table

Pretreating 10-d-old rats with Cr completely blocked electrographic and behavioral seizures during hypoxia (Table 2,Fig. 1). The high incidence of hypoxia-induced seizures in controls has been observed previously in 10-d-old rats(3). Behavioral seizures were clonic or myoclonic. Control animals showed rhythmic, high voltage spikes often associated with behavioral seizures (Fig. 1). Three Cr-treated 10-d-old pups showed increased spike activity, which was not rhythmic or sustained and was not associated with behavioral seizures. None of the 20-d-old rats showed electrographic or behavioral seizures.

Figure 1
figure 1

EEGs in 10-d-old rat pups acquired before and during a hypoxic period breathing 4% O2. The top tracing is from a rat pup which received s.c. Cr injections on d 7-9. No behavioral or EEG seizures were seen. The lower tracing is from a control pup which received saline on d 7-9. Electrocortical and clonic seizures were seen within 1 min of breathing the hypoxic gas mixture.

Full size image

Changes in NMR spectra during hypoxia were similar in control and Cr-injected rats at each age (Fig. 2). At 10 d, the PCr/NTP ratio decreased in parallel in the two groups, maintaining a constant difference until most control animals died at 4-6 min of the posthypoxic period. During hypoxia, PCr decreased 40-50% and ATP decreased 15% in both 10-d-old groups. This calculated ATP decrease assumes that the NTP loss during hypoxia was entirely ATP, which is 60-70% of the baseline total NTP(16,17). Brain pH decreased from 7.05 to 6.95 during hypoxia in both groups. In treated 10-d-old pups, the PCr/NTP ratio and each reactant increased within 2-4 min of rebreathing room air, whereas in controls further decreases in this ratio and reactants were seen(p < 0.05 at 4 min). Deaths occurred in the control pups within 4-6 min after beginning to breathe room again. In 20-d-old controls and Cr-treated rats, the PCr/NTP ratio decreased 50% during hypoxia. The PCr concentration decreased 60-70%. After an initial 15% loss, ATP decreased 70% in the last 4 min of hypoxia. These brain CK reactant changes are like those previously described in hypoxic 20-d-old rats(18).

Figure 2
figure 2

Changes in brain PCr/NTP ratio and in PCr and NTP concentrations during and after an 8-min hypoxic period in 10-d-old control rats (□-□) and rats which received s.c. Cr injections(·---·) as described in "Methods." Each value is the mean of results in seven rats. Concentration ratios and changes in concentrations were calculated from partially saturated NMR spectra. The PCr/NTP ratio was different before hypoxia with this difference maintained throughout the hypoxic period. The asterisk (*) indicates a significant difference(p < 0.05) between the groups during the posthypoxic period.

Full size image

DISCUSSION

In summary, Cr protects the rat pup, at a specific developmental stage, from hypoxic effects on survival and brain function. Hypoxic 10-d-old control rats show either a high incidence of seizures, if unanesthetized, or of death, if anesthetized. In treated pups respiratory efforts and heart rate persist probably due to prolongation of brainstem synaptic activity during hypoxia(19,20). Systemic injections of Cr increase the brain PCr/NTP concentration ratio measured in vivo by31 P NMR. Because brain NTP concentrations reportedly do not change with Cr loading, the PCr concentration probably is higher in brains of Cr-treated rats(20). Changes in brain Cr concentration are not known. Brain CK-catalyzed reaction rate constants are the same in control and Cr-injected pups, indicating a higher CK reaction rate in treated pups. In addition to the higher brain PCr/NTP ratio, the rapid recovery of high energy compounds after hypoxia in 10-d-old treated pups may contribute to hypoxic survival.

The rapid recovery of brain energy state after hypoxia in Cr-injected pups is consistent with the proposed central role of the CK system in closely coupling and regulating ATP and ADP metabolism in cerebral cortical gray matter(21–24). This interpretation implies that the immature CK system requires only the increase in substrate concentrations to achieve this function. Even though the pseudo-first order rate constant for the CK-catalyzed reaction does not change in the Cr-treated pups, isoenzyme activities and concentrations of other constituents of the CK complex may change with the developmentally early increase in brain CK reactant concentrations. Further studies of the CK complex in the developing brain are necessary.

The observation that PCr increases in mature cerebral gray matter during seizures plus the present demonstration of seizure prevention and increased brain PCr/NTP concentration ratio in the immature brain resulting from Cr injections support the hypothesis that Cr and/or PCr plays a role in limiting seizure activity in the mature brain(12). Concentrations of brain PCr and Cr both may increase with Cr injections, and both reactants must be considered in potential protective mechanisms. Measurement and localization of the two reactants after Cr injections will be important in testing proposed protective mechanisms. The loss of PCr during hypoxia may be critical because of the low Cr and PCr concentrations and increased excitability characteristic of the immature brain(1,4). Increasing Cr also may protect the immature heart during hypoxia and secondarily protect the brain with more stable cerebral perfusion.

These results have important implications for the developmental mental physiology and pathophysiology of brain energy metabolism in species with postnatal metabolic development, including the human. Similar postnatal increases in PCr/NTP ratios in the human and rat brain suggest that the rodent is an appropriate model for studying development of brain ATP metabolism. The association of low PCr/NTP ratio and low seizure thresholds indicates a comparable developmental stage in the 10-12-d-old rat and the human term newborn. Suppression of hypoxia-induced seizures by Cr and the increase in PCr during seizures in the mature cerebral gray matter suggest that Cr and/or PCr limit seizure susceptibility. Whether the protection of the rat pup during hypoxia is due to this effect of Cr on seizures or whether Cr/PCr facilitates ATP metabolic coupling during energy deficit states and secondarily suppresses hypoxic seizures is uncertain. Protection of brainstem control of respiratory and cardiac functions also may contribute to hypoxic survival in the Cr-treated rat pups. The protective effects of Cr in the immature brain may share a mechanism with the recently described protective effects of Cr in an animal model of Huntington's disease(25). An important physiologic function for Cr also is supported by the facts that Cr, a compound available in normal diet, is synthesized and transported in brain(26–28). The postnatal altricial rodent may be used for studies of Cr and its analogs in brain protection(29). The Cr treatment of rat pups has not been studied systematically to optimize dose and treatment schedules. This initial study suggests that increasing brain Cr/PCr in the metabolically immature brain by systemic Cr administration could be an important method of protecting cell viability and function.