| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
MS 6843 Received 21 April 1997; accepted after revision 2 September 1997.
 |
ABSTRACT |
|---|
 TopAbstract IntroductionMethodsResultsDiscussionReferences |
|---|
|
INTRODUCTION |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Prostaglandins (PGs) have neuromodulatory effects in the central and peripheral nervous systems and are known to influence catecholamine and cholinergic transmission (Hedqvist, 1977), and the release of several neuropeptides, especially those synthesized and released by the hypothalamus and pituitary gland. At physiological pH, PGs do not readily cross cell membranes (Bito & Baroody, 1975; Bito & Wallenstein, 1977) and therefore, it is likely that PGs do not enter the brain from the circulation in physiologically significant amounts. Rather, it has been shown in adult animals that PGs are cleared from the brain into the circulation by a saturable, energy-dependent transport mechanism (Bito, Davson & Hollingsworth, 1976). Whether such a mechanism is present in the fetal brain, and acts to limit the penetration of PGs from the circulation into the brain is not known. Cerebral tissue has little capacity to metabolize PGs (see Coceani, 1974), and the PGs found in cerebral extracellular fluid (ECF) and cisternal cerebrospinal fluid (CSF) probably reflects the endogenous synthesis of prostanoids by the brain.
Koos (1985) found that inhibition of brain PG synthesis in fetal sheep by ventriculo-cisternal perfusion of meclofenamate resulted in a marked increase of breathing movements, suggesting that cerebral synthesis of PGs is important in regulating respiratory activity in utero. PG synthase has been localized immunocytochemically throughout the brain of adult sheep (Breder et al. 1992) and in the medulla of fetal sheep at 126 days gestation (Norton, Smith, Adamson, Bocking & Han, 1990). Attempts to selectively suppress the cerebral synthesis of PGs by infusion of indomethacin, a cyclo-oxygenase inhibitor, into the third ventricle of fetal sheep have only been partially successful, because plasma concentrations of PGE2 were also decreased by this treatment (Jones, Adamson, Bishai, Lees, Engelberts & Coceani, 1994), albeit to a lesser extent than that which occurred in the CSF. However, in these experiments, and those of Koos (1985), it is likely that the concentration gradient of PGs between the blood and brain was increased after the central inhibition of PG synthesis, even if some suppression of peripheral synthesis had also occurred. Thus, these studies suggest that PGs do not readily pass from the fetal circulation into the brain.
The transport of a number of substances from the brain into blood occurs at both the cerebral endothelial interface (i.e. the 'blood-brain barrier') and the choroid plexus. The ability of probenecid to inhibit the transport of PGs by the isolated choroid plexus, and from CSF to blood in the intact blood-brain barrier has been well described (Bito, 1975; Bito, Davson & Hollingsworth, 1976). If intracerebral synthesis of PGE2 is important in the regulation of the activity of the respiratory centres in fetal sheep, it should be possible to show that probenecid treatment will lead to an increase of PGE2 concentration in CSF, and a partial or complete inhibition of spontaneous breathing movements. Unanaesthetized fetuses were thus treated in utero with probenecid either, while breathing movements were being recorded or, while cisternal CSF was being collected from a catheter chronically implanted in the cisterna magna. In addition, anaesthetized, exteriorized fetuses were studied to directly determine the clearance of [3H]PGE2 from the brain using a ventriculo-cisternal perfusion technique. The results indicate that PGE2 is normally transported out of the brain into the bloodstream in the fetus, as in the adult, and that intracerebral synthesis of PGE2 participates in the regulation of fetal respiratory activity.
|
METHODS |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Twenty four pregnant sheep were used in this study. All procedures had received prior approval from the Standing Committee on Ethics in Animal Experimentation at Monash University.
Studies on unanaesthetized fetuses, in utero
Nineteen ewes (117-123 days gestation) underwent a laparotomy while anaesthetized with halothane (1-2 %) and nitrous oxide- oxygen (50 : 50 vol/vol) given through an endotracheal tube. Sterile operating procedures were used throughout. The fetal head and neck were withdrawn through an incision in the uterus, and catheters directed towards the heart were inserted into a carotid artery, jugular vein and the trachea. After closing the skin, an open-ended catheter was sewn to the skin on the back of the neck to measure amniotic fluid pressure. In fifteen of the fetuses, insulated stainless steel wire electrodes (Cooner Wire, Chatsworth, CA, USA) were used to record electrocortical (ECoG) activity. The electrodes were placed bilaterally onto the dura over the parietal cortex through 1 mm holes drilled in the skull, and held in place with a drop of cyano-acrylate glue. In another four fetuses a Silastic catheter (o.d., 1·7 mm; i.d., 0·8 mm; Dow Corning, MI, USA) was inserted through the dura directly beneath the atlanto-occipital membrane; the dura was punctured with the tip of an 18 gauge needle and the catheter was inserted approximately 1 cm rostrally toward the cisterna magna. The catheter was held in place by 5/0 monofilament suture tied around the catheter and passed through the edges of the atlanto-occipital membrane (ECoG and EMG electrodes were not attached in these fetuses). After repairing the skin incisions, each fetus was given a 2 ml intramuscular injection of Depomycin (Intervet, Castle Hill, NSW, Australia) containing procaine penicillin (400 mg) and dihydrostreptomycin sulphate (500 mg) and replaced in the uterus. The incisions in the membranes, uterus, abdominal wall and skin of the ewe were then repaired, and the anaesthetic gases were withdrawn and the ewe allowed to recover. Experiments began after allowing at least 5 days for recovery from surgery.
Fetal breathing movements were recorded on a polygraph as the phasic changes of tracheal pressure, after electronic subtraction of amniotic pressure to account for the pressure changes due to movements of the ewe relative to the fixed position of the pressure transducers. Fetal blood pressure and heart rate were also recorded. ECoG activity was recorded using a high gain, differential input amplifier (7P2B, Grass Instrument Co.) and displayed on the polygraph after passing the signal through a bandpass filter (1-35 Hz).
On the day of an experiment, fetal breathing movements and ECoG activity were recorded for at least 3 h before commencing an infusion of probenecid into the fetal jugular vein catheter. Seven fetuses received probenecid (
100 mg kg-1) alone, followed on another day by infusion of the probenecid vehicle infusion. Probenecid (Sigma) was made up in a minimal volume of 1 M NaOH until dissolved, and was then diluted in saline, and the pH adjusted to approximately 8·0 using HCl. The dose was infused in a volume of 10-15 ml into the fetal jugular vein over 1 h. The expected fetal weight at the particular gestational age was used to calculate the amount of probenecid that should be infused; the actual dose given was subsequently calculated from a fetal weight estimated from the post-mortem weight, using the equation given by Lumbers, Smith & Stevens (1985). The polygraph recordings continued for at least 6 h after the probenecid infusion. Arterial blood samples were taken at regular intervals for measurement of blood gases and pH using an ABL30 blood gas analyser (Radiometer, Copenhagen), and for radioimmunoassay of plasma PGE2 as described by Fowden, Harding, Ralph & Thorburn (1987).
Ten fetuses received an infusion of paracetamol to inhibit PG synthase activity; eight of these fetuses had not previously received any treatment, and two fetuses had received the probenecid/vehicle treatments described above at least 3 days beforehand. Ten millilitres of paracetamol solution (21 mg ml-1) was infused into the jugular vein over 10 min, followed by infusion of a further 18-24 ml at 6 ml h-1. Probenecid was then infused when the fetal breathing movements had been clearly increased in both incidence and amplitude for at least 45 min. The actual time between starting the paracetamol infusion and the onset of the probenecid infusion was 2-4·5 h in the different experiments. In six of the experiments, probenecid was infused after the administration of paracetamol had been completed, and in four experiments the infusion of probenecid began while the paracetamol infusion was still in progress.
At 129-133 days gestation CSF was collected from the four fetuses in which the cisternal catheter had been implanted. CSF was allowed to drain by gravity from the catheter when the distal end was held below the level of the ewe's abdomen. The CSF drained into pre-weighed 1·5 ml Eppendorff tubes inserted into a specially designed holder attached to the side of the cage. Samples were collected over 30 min periods, and the volume determined by re-weighing each tube. The samples were then immediately frozen in liquid N2 and stored at -70°C until analysis of PGE2. After 4 × 30 min samples had been collected a 1 h infusion of probenecid (100 mg ml-1) was commenced. CSF was collected as 30 min samples over the next 6 h. Two fetuses received the probenecid infusion first, followed by the vehicle infusion 2 days later, and the other two fetuses received the treatments in the reverse order. An arterial blood sample (1·5 ml) was taken hourly for the measurement of plasma PGE2 concentrations. Three of these fetuses were then treated with paracetamol as above, and CSF was collected over 5 h (n = 1) and 8 h (n = 2) starting 2 h after the onset of the paracetamol infusion.
Exteriorized fetuses
Five fetuses (128-140 days gestation) were studied while the ewe was anaesthetized by intermittent injection of pentobarbitone (3-6 mg kg-1; Nembutal, Boehringer Ingelheim, Artamon, NSW, Australia) given through a catheter inserted percutaneously into a jugular vein. The ewe was then ventilated via an endotracheal tube, and given O2 to maintain fetal and maternal blood gases in their normal range, if necessary. The ewe was placed on her left side, the abdomen incised in the mid-line, and the part of the uterus containing the fetal head was withdrawn and placed inside a large plastic bag to prevent loss of moisture by evaporation. The plastic and the uterus were incised and the fetal head and neck withdrawn. Catheters were placed in a carotid artery and jugular vein, and the head positioned so that a catheter (o.d., 0·96 mm; i.d., 0·58 mm; SV45, Dural Plastics, NSW, Australia) could be placed in the cisterna magna through the dura beneath the atlanto-occipital membrane. The catheter was pushed approximately 1 cm rostrally through a hole in the dura made by the tip of an 18 gauge needle. The catheter was manipulated until CSF flowed out freely when the other end was held at or slightly below the level of the fetal head. A cannula made from stainless steel tubing (o.d., 1 mm; i.d., 0·56 mm) was then inserted into a lateral ventricle through a 1 mm hole drilled in the skull 3 mm from the mid-line and 5 mm rostral to the coronal suture. The inward flow of saline under gravity through a catheter attached to the cannula was used to indicate when the tip of the probe had passed through the brain into the ventricle. The cannulae were 12-16 mm in length, depending on the size of the fetal head, and were fixed to the skull by a drop of cyano-acrylate glue. Warmed artificial CSF made up as described by Bissonnette, Hohimer & Richardson (1981), and containing sufficient Blue Dextran (Pharmacia, Uppsala, Sweden) to allow the colour of samples to be read directly at 620 nm by a colorimeter, was then infused into the ventricular cannula at 104 µl min-1 using a mini-roller pump (Miniperpex, LKB, Bromma, Sweden). CSF was collected from the cisternal catheter over 5 min intervals using a fraction collector (Pharmacia, Uppsala, Sweden). The end of the outflow catheter was held at the height of the fetal head to maintain normal pressure in the ventricular system. When the cisternal outflow showed a blue colour, indicating that the ventricular system was being perfused successfully, a sample was collected over 10-15 mins for measurement of PGE2 concentration, and an arterial blood sample was also taken at this time. Then, 60 × 103-500 × 103 d.p.m. of [3H]PGE2 was added to the inflow reservoir (corresponding to 0·5 × 10-15-4 × 10-15 M PGE2, NEN-Dupont). Between 1·5 and 2 h after starting the ventriculo-cisternal perfusion, probenecid was then infused into the jugular vein and the ventriculo-cisternal perfusion continued for a further 1-2 h. The dose and infusion rate of probenecid was as for the experiments on the unanaesthetized fetuses. Arterial blood samples were taken from the ewe and fetus at intervals for measurement of blood gases and pH. The exposed parts of the uterus and fetus were wrapped in cotton wool and towelling to minimize heat loss. Anaesthesia was maintained throughout the 4-6 h of the experiment by regular injections of pentobarbitone (3-6 mg kg-1) into the jugular catheter in the ewe. At the end of the experiment the ewe and fetus were killed by intravenous injection of 20 ml of concentrated pentobarbitone (325 mg ml-1; Lethobarb, Virbac, Peachill, NSW, Australia) into the ewe.
Analyses
Fetal breathing movements were analysed from the polygraph records using a digitizing tablet and pen connected to a minicomputer. Each record was divided into 1 h epochs, and the amplitudes of the tracheal pressure deflections were measured at 1 min intervals. A data file showing the number of minutes in each epoch that breathing movements were present, and their mean amplitude, was computed. Tracheal pressure changes less than 1 mmHg were treated by the computer as zero values. The ECoG record was divided into episodes of high and low voltage activities by visual analysis and the number of minutes of each type of activity tabulated for each hour of the record.
Plasma and CSF PGE2 concentrations were measured as the PGE-methyloxime derivative by direct radioimmunoassay (Fowden et al. 1987), using antisera raised in sheep. The sensitivity of the assay was 0·05 nmol l-1. The intra-assay coefficient of variation was < 10 %, and interassay coefficients of variation were 10 % at 14·58 nmol l-1 and 15·6 % at 43·24 nmol l-1. Fetal and maternal blood gases were measured at 37°C and corrected to 39°C and 38°C, respectively. [3H]PGE2 in CSF was counted in a Beckman liquid scintillation counter, using a mixture of 100 µl CSF and 1·5 ml of Ecoscint (National Diagnostics, Atlanta, GA., USA). When CSF samples were incubated with the methoximating reagent overnight at room temperature, and then with methyloxamine-PGE2 antiserum, all 3H activity was precipitated by addition of polyethylene glycol, indicating that the 3H activity in the sample was still associated with PGE2 and that negligible breakdown of [3H]PGE2 had occurred during the perfusion through the fetal brain. Dextran in CSF was measured directly in a 200 µl-1 sample using a colorimeter with the incident light filter set at 620 nm.
Clearance of a substance from CSF, and the production rate of CSF was calculated as described by Heisey, Held & Papperheimer (1962) and Bito, Davson & Hollingsworth (1976). Clearance of [3H]PGE2 from CSF was calculated as:
where Cin and Cout refer to the activity of [3H]PGE2 in the infusate and cisternal CSF, and C 'in and C 'out refer to the concentration of Dextran in the infusate and cisternal CSF, respectively. F is the rate of perfusion (104 µl min-1). CSF production rate (V) was calculated as:
where, C 'in and C 'out refer to the concentration of Dextran in the infusate and cisternal fluids, respectively.
Statistics
Data are presented as means ±
Unanaesthetized fetuses
Breathing movements. Intravenous infusion of probenecid was associated with a reduction in the incidence of fetal breathing movements as shown in Fig. 1. In eighteen trials on seven fetuses the incidence of breathing movements decreased from a mean control level of 29·0 ± 1·6 min h-1 to 16·6 ± 2·4 min h-1 during the hour of the infusion (P < 0·05), and to 20·8 ± 1·9 min h-1 during the hour immediately after the infusion (P < 0·05; Fig. 2A). Two hours after the end of the infusion the incidence of breathing movements was not significantly different from the control value. The amplitude of the breathing movements decreased significantly from a mean control level of 3·3 ± 0·2 to 2·5 ± 0·2 mmHg during the infusion (P < 0·05) and was 3·0 ± 0·2 mmHg in the hour after the probenecid infusion (Fig. 2B). Infusion of the vehicle solution (10 trials, 6 fetuses) at the same rate as for probenecid had no significant effect on the incidence or amplitude of the breathing movements (Fig. 2A and B). Probenecid infusion did not change the incidence of low voltage ECoG activity, plasma concentrations of PGE2, blood gases, pH, mean arterial pressure and heart rate, which all were in their normal range (Table 1).
Figure 1. Breathing movement in a fetus at 131 days gestation, shown as the downward (negative) deflections of tracheal pressure
The three traces are sequential; numbers above each trace show the time in hours with respect to the start of the
Figure 2. Effects of intravenous infusion of probenecid to the fetus
Mean incidence (A, min h-1) and mean amplitude (B, mmHg) of fetal breathing movements before, during and after intravenous infusion of probenecid (continuous lines; 18 trials, 7 fetuses) or vehicle (dashed lines; 10 trials, 6 fetuses). Thick bar shows the time of infusion. * Significant differences from control values within treatment groups (P < 0·05). Data shown as means ±
Table 1. Incidence of low voltage ECoG activity, plasma PGE2 concentrations, arterial blood gases and pH, mean arterial pressure and heart rate in fetal sheep before and at the end of a 1 h infusion of probenecid
To show that the effect of probenecid on breathing movements was dependent on PG synthesis, ten fetuses were first given an intravenous infusion of the PG synthase inhibitor paracetamol. The plasma concentrations of PGE2 decreased from 3·02 ± 0·32 to 0·86 ± 0·11 nM (P < 0·01) at 5 h after beginning the infusion (Fig. 3). As shown for other PG synthase inhibitors such as indomethacin (Kitterman, Liggins, Clements & Tooley, 1979) and meclofenamate (Wallen, Murai, Clyman, Lee, Mauray & Kitterman, 1986), the incidence of breathing movements increased significantly from a mean control value of 29·2 ± 4·5 to 53·3 ± 1·7 min h-1 (P < 0·01), and the amplitude increased significantly from 3·8 ± 0·4 mmHg to a peak of 12·2 ± 1·4 mmHg (P < 0·01). Intravenous infusion of probenecid when the breathing movements were augmented had no effect on either the incidence or amplitude (Fig. 4). The mean data for the ten fetuses which received paracetamol and then probenecid is shown Fig. 5. There was no significant change of blood gases, pH, mean arterial pressure or heart rate during the infusion of paracetamol or during the subsequent infusion of probenecid.
Figure 3. Plasma concentration of PGE2 with paracetamol
The plasma concentration of PGE2 (nm) in 10 fetuses which received an intravenous infusion of paracetamol. Paracetamol was present in all experiments from 2 h before infusion of probenecid (thick bar with arrow); in some experiments paracetamol infusion began before -2 h (see Methods).
Figure 4. Breathing movements in a fetus at 135 days gestation after i.v infusion of paracetamol
The two traces are sequential; numbers above each trace show the time in hours with respect to the start of the
Figure 5. Mean incidence of fetal breathing movements before, during and after
Paracetamol infusions began -4·5 to -2 h before the onset of probenecid infusion (thick bar). Data shown as means ±
CSF PGE2. The concentration of PGE2 in cisternal CSF was between 2·7 and 4·6 nM in the different fetuses (mean, 4·3 ± 0·8 nM) immediately before either the probenecid or vehicle solutions were infused. Infusion of probenecid resulted in a 4·8- to 11·2-fold increase of the PGE2 concentrations (mean, 6·6 ± 1·5-fold; P < 0·05) observed 2·5 h after the start of the infusion (Fig. 6). The delay between the time the probenecid was infused and the increase of PGE2 concentration must have been due in part to the relatively slow rate at which CSF drained (11-40 µl min-1), and the dead-space of the catheter which was approximately 400 µl. The flow of CSF was not affected by the infusion of either probenecid or the vehicle solution, and did not change systematically over the 8 h of the experiment. In the control experiments the PGE2 concentrations did not increase after the infusion, but decreased slightly after 6 h of drainage (i.e. 4 h after the start of the infusion; Fig. 6). This has been observed in other experiments where CSF has been drained for long periods of time (D. W. Walker & N. Pratt, unpublished observations), and may be due to the dilution of PGE2 by newly formed CSF. In 3/4 fetuses which received paracetamol 2 h before starting to drain CSF the concentrations of PGE2 were below the sensitivity of the assay (0·05 nM) during 5 h (n = 1) and 8 h (n = 2) of CSF drainage.
Figure 6. PGE2 concentrations in cisternal CSF
The PGE2 concentrations are expressed as a percentage of the concentration in the sample obtained immediately before the start of an infusion of probenecid (
Anaesthetized fetuses
Five fetuses (128-140 days) were exteriorized while the ewe was maintained under pentobarbitone anaesthesia. The resting PGE2 concentration of cisternal CSF collected from the five exteriorized fetuses after the ventriculo-cisternal perfusion had started, but before the [3H]PGE2 was added, was 0·65 ± 0·08 nM. The plasma concentration at this time was 3·11 ± 0·72 nM, significantly higher than the CSF concentration of PGE2 (P < 0·05). Perfusion with mock CSF containing Blue Dextran and [3H]PGE2 resulted in an outflow concentration and activity of Dextran and [3H]PGE2 which were less than the respective inflow concentration and activity, but which reached a steady state after 2-3 h (Fig. 7). The mean outflow concentration of Dextran relative to the inflow concentration was 69·5 ± 4·1 %. The mean CSF production rate was calculated to be 47·5 ± 5·1 µl min-1 (Table 2). The outflow activity of [3H]PGE2 relative to the inflow activity was 13·4 ± 4·2 %, which was significantly less than for Dextran (P < 0·05). The steady-state clearance of [3H]PGE2 from the perfusate was 151·6 ± 15·1 µl min-1 (Table 2).
Figure 7. Concentration of Dextran
The concentration of Dextran (mg ml-1;
Figure 8. [3H]PGE2 activity
Activity of [3H]PGE2 (d.p.m. (0·1 ml)-1) in cisternal cerebrospinal fluid during ventriculo-cisternal perfusion of a fetus, 132 days gestation. [3H]PGE2 was added to the mock CSF at the time shown by the arrow and probenecid was infused intravenously during the time shown by the thick bar.
Table 2. Infusion of probenecid
These experiments show that when probenecid is infused into fetal sheep there is a decrease in the incidence and amplitude of spontaneous breathing movements, an increase of PGE2 concentration in cisternal CSF, and a decrease in the rate at which PGE2 is lost from the ventricular system of the brain. These results are interpreted as showing that there is active transport of PGs from the brain to the circulation in fetal sheep, and that inhibition of this transport with probenecid results in the accumulation of PGs within the brain, leading to the suppression of activity of the brainstem respiratory centres.
Other conditions which inhibit breathing movements in fetal sheep include hypoxia (Boddy, Dawes, Fisher, Pinter & Robinson, 1974) and increased plasma concentrations of PGE2 (Wallen et al. 1986), but neither the fetal arterial PO2 nor plasma PGE2 concentrations were changed by the infusion of probenecid. The incidence of breathing movements would also have decreased if probenecid had decreased the incidence of low voltage ECoG activity, because breathing movements are normally present only during this ECoG activity state (Dawes, Fox, Leduc, Liggins & Richards, 1972). However, the decrease in the incidence of breathing movements occurred independently of any change in the overt pattern of ECoG activity. Probenecid also inhibits the clearance of some catecholamine and indoleamine metabolites (Ashcroft, Dow & Moir, 1968; Tamarkin, Goodwin & Axelrod, 1970; Anderson & Roos, 1972), of cyclic AMP (Cramer & Lindl, 1972; Extein, Korf, Roth & Bowers, 1973) and of the NMDA/glycine receptor antagonist kynurenic acid from the CSF of adult animals (Stone & Connick, 1985). Evidence suggests that centrally released monoamines augment rather than decrease breathing movements in fetal sheep (Joseph & Walker, 1990, 1993), but it is not known if any of the catecholaminergic metabolites have an inhibitory effect on respiratory activity in the fetus. Kynurenic acid is present in CSF of fetal sheep and also increases after probenecid treatment (B. Curtis & D. W. Walker, unpublished observations), and this could have contributed to the reduction of breathing movements. However, the observation that probenecid did not have an effect on breathing movements after inhibition of PG synthase activity with paracetamol suggests that (a) active synthesis of PGs is necessary for probenecid to decrease the incidence of breathing movements, and (b) the effect is not due to another, perhaps non-specific, action of probenecid. Intraperitoneal administration of probenecid in adult rabbits causes a gradual increase of cyclic AMP in CSF over 3-5 h (Sebens & Korf, 1975), but the more rapid effect of probenecid on breathing movements, and its reversal, that we observed suggests that the suppression of fetal respiratory activity occurs before the significant accumulation of a second messenger or metabolite.
The increase of breathing movements which occurred after paracetamol was similar to that reported for other cyclooxygenase inhibitors such as indomethacin and meclofenamate (Kitterman et al. 1979; Wallen et al. 1986). The increase of respiratory activity after inhibition of PG synthesis in the fetus is not fully understood. PGE2 binding sites are present in fetal sheep brain throughout the medullary and pontine regions concerned with cardiorespiratory control (Tze-Chuan, MacLusky & Adamson, 1994). Constitutive and inducible forms of cyclo-oxygenase are widely distributed throughout the brain, but occur predominantly in the cell body and dendritic region of neurones (Breder et al. 1992; Breder, Dewitt & Kraig, 1995). The presence of PG synthases in the cell body rather than axonal terminals suggests that prostanoids may function to regulate the effects of input into the neurone; i.e. to regulate the excitability of the neurone. Thus, prostanoids may account for the relatively weak cardiorespiratory reflexes which are elicited from the peripheral and central chemoreceptors in the fetus, compared with the newborn and adult sheep (reviewed by Guissani, Spencer & Hanson, 1994). Cyclo-oxygenase inhibitors such as indomethacin also reduce cerebral blood flow in fetal sheep (van Bel, Bartelds, Teital & Rudolph, 1995), which could conceivably affect the activity of the central chemoreceptors. Although a clear account of why fetal breathing movements increase after inhibition of PG synthesis with drugs such as paracetamol cannot be given, the results of this study show, nevertheless, that PG synthesis was necessary for probenecid to produce a suppression of respiratory activity. The most reasonable explanation for this effect is that inhibition of organic acid transport across both the cerebral endothelial (i.e. the blood-brain barrier) and the choroid plexuses caused the accumulation of PGs within the brain. It is possible that probenecid also inhibited the transfer of PGs from the circulation into the brain since most transport processes are potentially bidirectional. However, this could not explain the effect of probenecid on fetal breathing movements because, if the primary contribution of PGs in the brain were from the circulation, probenecid would have decreased cerebral PG concentrations and breathing movements would have increased, as they do after inhibition of PG synthesis. The possibility that breathing movements decreased because PGs entered the brain more rapidly after probenecid treatment is not likely, but was not tested in the present experiments.
In the anaesthetized fetuses the resting concentrations of PGE2 in cisternal CSF, collected immediately after the ventriculo-cisternal perfusion with mock CSF had started but before the [3H]PGE2 tracer had been added, were significantly lower than the plasma concentrations of PGE2. Whether this concentration difference normally exists in utero is unclear (see below), but this result shows that a considerable concentration difference can occur between blood and CSF, and is consistent with the concept that the cerebral endothelial interface and the choroid plexus is relatively impermeable to PGs. The low concentration of PGE2 in CSF in these experiments may have been due to dilution because of the perfusion, or perhaps due to a decrease in the synthesis of PGE2 because of the pentobarbitone anaesthesia, although the systemic production of PGE2 was apparently not affected because the plasma concentrations were in the normal range for fetal sheep at this stage of gestation. In the chronically catheterized fetuses the CSF PGE2 concentrations were considerably higher, and similar to those of plasma. Other studies have reported cisternal CSF PGE2 concentrations in chronically catheterized fetal sheep equal to, and even higher than plasma concentrations (Smith, Brien, Homan, Carmichael & Patrick, 1990), although it is notable that in one study the PGE2 concentrations measured in third ventricular CSF (366 ± 120 pg ml-1) were consistently less than the plasma concentrations of PGE2 (520 ± 69 pg ml-1; Jones et al. 1993). Notwithstanding the possibility that chronically indwelling catheters are irritant and cause increased PG synthesis, or that there are significant differences in the concentration of PGE2 between the different ventricular spaces in the brain, the present study shows that (a) a considerable concentration difference can exist between CSF and plasma, (b) inhibition of epithelial transport with probenecid increased PGE2 concentrations in CSF of the fetus in utero, and (c) probenecid decreased the clearance of PGE2 from the brain in the anaesthetized, exteriorized fetus. These results are consistent with the conclusion that the movement of eicosanoids into the brain from the circulatory system is restricted, and that there is a net efflux of these substance from the brain into the circulatory system.
The CSF production rate in the anaesthetized fetuses was within the range of values reported by others (21·4-77·0 l min-1) for fetal sheep at similar gestational ages whether unanaesthetized and in utero (Bissonnette, Hohimer & Richardson, 1981) or anaesthetized and exteriorized from the uterus (Fossan et al. 1985). During ventriculo-cisternal perfusion the outflow concentration of Dextran was less than the inflow concentration, attributable to dilution of Dextran by both the existing CSF volume and the production of new CSF (Heisey et al. 1962; Bito et al. 1976). In comparison, the outflow of [3H]PGE2 in relation to its inflow activity was significantly less than could be accounted for by dilution in CSF, indicating that some of the tracer had been removed from the ventricular spaces. The [3H]PGE2 could have diffused into the brain from the ventricles, but this would result in an eventual equilibration between CSF and brain extracellular fluid, and therefore an equilibration of outflow concentration of [3H]PGE2 close to that expected from dilution in CSF alone. This did not occur even after several hours of ventriculo-cisternal perfusion. Some of the [3H]PGE2 may have passed into the blood stream at those areas in the brain where exchange of small molecules between the blood and brain is relatively unrestricted (e.g. area postrema, organ vasculosum lamina terminalis (OVLT)). At physiological pH, it is likely that PGs do not readily cross cell membranes (Bito & Baroody, 1975) and unless transfer occurred through endothelial pores this process is still likely to involve transcellular diffusion.
The [3H]PGE2 clearances measured in this study were of an order of magnitude greater than those reported by Bito et al. (1976) for the adult rabbit brain. This may be a species difference, or represent a quantitative difference reflecting the different physiological states which exist in fetal and adult life. In fact, the volume of CSF cleared of [3H]PGE2 was nearly equal to the rate of infusion of mock CSF plus the estimated rate of CSF production. The high clearance was not apparently due to the breakdown of [3H]PGE2 in CSF because the [3H]activity in the cisternal outflow was immunoprecipitated by PGE2 antibody. Some of the [3H]PGE2 may have entered the interstitium of the brain, but the activity in brain tissue relative to CSF (i.e. the [3H]PGE2 'space') was not measured in these experiments to test this possibility.
The results thus indicate that PGE2 accumulates within the fetal brain when transport across the blood-brain barrier and choroid plexuses is blocked with probenecid. The increased concentrations result in suppression of respiratory activity, suggesting that intracerebral synthesis of PGE2 may be important in determining the activity of the respiratory centres before birth. Whether the clearance of PGE2 from the brain normally occurs down or against a concentration gradient in the fetus requires further work to establish the true concentration difference for PGE2 between the brain and blood.
Anderson, H. & Roos, B.-E. (1972). 5-Hydroxyindole acetic acid and homovanillic acid in cerebrospinal fluid and brain of different rabbit breeds after treatment with probenecid. Journal of Pharmaceutics and Pharmacology 24, 165-166.
This project was supported by a grant from the National Health and Medical Research Council of Australia. We are grateful to Roseanne Gleeson and Brenda Lee for help with the experiments, and to Alex Satragno and Cheryl McKechnie for assistance with surgery.
Corresponding author
D. W. Walker: Department of Physiology, Monash University, Clayton, Victoria, 3168, Australia.
Email: d.walker{at}med.monash.edu.au
This article has been cited by other articles:
RESULTS
Top
Abstract
Introduction
Methods
Results
Discussion
References
View larger version
[in this window]
[in a new window]
View larger version
[in this window]
[in a new window]
Low voltage ECoG (min h-1) PGE2 (nM) PO2 (mmHg) PCO2 (mmHg) pH O2 sat (%) Mean arterial pressure (mmHg) Heart rate (beats min-1) Control 31·4 ± 2·3 4·90 ± 1·2 23·7 ± 1·0 50·5 ± 1·1 7·363 ± 0·007 62·9 ± 3·4 45·3 ± 1·6 147 ± 8 Probenecid 36·4 ± 2·2 4·53 ± 1·2 23·4 ± 0·8 49·4 ± 1·1 7·373 ± 0·005 64·7 ± 3·3 46·7 ± 2·1 151 ± 10
View larger version
[in this window]
[in a new window]
View larger version
[in this window]
[in a new window]
View larger version
[in this window]
[in a new window]
View larger version
[in this window]
[in a new window]
) or vehicle solution (
). The period of the infusion is shown by the thick bar. *P < 0·05, between groups comparison.
View larger version
[in this window]
[in a new window]
), and activity of [3H]PGE2 (d.p.m.; ) in cisternal cerebrospinal fluid, expressed as the percentage of the inflow concentration (or activity) during ventriculo-cisternal perfusion of a fetus, 128 days gestation, with mock CSF. [3H]PGE2 was added to the inflow reservoir at the time shown by the arrow.
View larger version
[in this window]
[in a new window]
CSF production (µl min-1) [3H]PGE2 clearance (µl min-1) Control 47·5 ± 5·1 151·6 ± 15·1 Probenecid 43·5 ± 5·9 126·1 ± 13·9 *
DISCUSSION
Top
Abstract
Introduction
Methods
Results
Discussion
References
REFERENCES
Top
Abstract
Introduction
Methods
Results
Discussion
References
Acknowledgements
Ashcroft, G. W., Dow, R. C. & Moir, A. J. B. (1968). The active transport of 5-hydroxyindol-3-ylacetic acid and 3-methoxy-4-hydroxy-phenylacetic acid from a recirculatory perfusion system of the cerebral ventricles of the unanaesthetized dog. The Journal of Physiology 199, 397-425
[Medline]
Bissonnette, J. M., Hohimer, A. R. & Richardson, B. S. (1981). Ventriculocisternal cerebrospinal perfusion in unanaesthetized fetal lambs. Journal of Applied Physiology 50, 880-883
[Medline]
Bito, L. Z. (1975). Saturable, energy-dependent, transmembrane transport of prostaglandins against concentration gradients. Nature 256, 134-136
[Medline]
Bito, L. Z. & Baroody, R. (1975). The impermeability of rabbit erythrocytes to prostaglandins. American Journal of Physiology 229, 1580-1584
[Medline]
Bito, L. Z., Davson, H. & Hollingsworth, J. R. (1976). Facilitated transport of prostaglandins across the blood-cerebrospinal fluid and blood-brain barriers. The Journal of Physiology 256, 273-285.
Bito, L. Z. & Wallenstein, M. C. (1977). Transport of prostaglandins across the blood-aqueous barriers and the physiological significance of these absorptive transport processes. Experimental Eye Research , suppl. 25, 229-243.
Boddy, K., Dawes, G. S., Fisher, R., Pinter, S. & Robinson, J. S. (1974). Foetal respiratory movements, electrocortical and cardiovascular responses to hypoxaemia and hypercapnia in sheep. The Journal of Physiology 243, 599-618
[Medline]
Breder, C. D., Dewitt, D. & Kraig, R. P. (1995). Characterization of inducible cyclooxygenase in the rat brain. Journal of Comparative Neurology 355, 296-315
[Medline]
Breder, C. D., Smith, W. L., Raz, A., Masferrer, J., Seibert, K., Needleman, P. & Saper, C. B. (1992). Distribution and characterization of cyclooxygenase immunoreactivity in the ovine brain. Journal of Comparative Neurology 322, 409-438
[Medline]
Coceani, F. (1974). Prostaglandins and the central nervous system. Archives of Internal Medicine 133, 119-124
[Medline]
Cramer, H. & Lindl, T. (1972). Probenecid inhibits efflux of adenosine 3'-5' monophosphate (AMP) from cerebrospinal fluid (CSF) in the rat. Psychopharmacologia 26, 29-33.
[Medline]
Dawes, G. S., Fox, H. E., Leduc, B. M., Liggins, G. C. & Richards, R. T. (1972). Respiratory movements and rapid eye movement sleep in the foetal lamb. The Journal of Physiology 220, 119-143
[Medline]
Extein, I., Korf, J., Roth, R. H. & Bowers, M. B. (1973). Accumulation of 3-methoxy-4-hydroxyphenylglycol sulphate in rabbit cerebrospinal fluid following probenecid. Brain Research 54, 403-407
[Medline]
Fossan, G., Cavanagh, M. E., Evans, C. A. N., Malinowska, D. H., Mollgard, K., Reynolds, M. L. & Saunders, N. R. (1985). CSF-brain permeability in the immature sheep fetus: a CSF-brain barrier. Developmental Brain Research 18, 113-124.
Fowden, A. L., Harding, R., Ralph, M. M. & Thorburn, G. D. (1987). The nutritional regulation of plasma prostaglandin E concentrations in the fetus and pregnant ewe in late gestation. The Journal of Physiology 394, 1-42
[Abstract]
Guissani, D. A., Spencer, J. A. D. & Hanson, M. A. (1994). Fetal cardiovascular responses to hypoxaemia. Fetal and Maternal Medicine Review 6, 17-37.
Hedqvist, P. (1977). Basic mechanism of prostaglandin actions on autonomic transmission. Annual Review of Pharmacology and Toxicology 17, 259-279
[Medline]
Heisey, S. R., Held, D. & Pappenheimer, J. R. (1962). Bulk flow and diffusion in the cerebrospinal fluid system of the goat. American Journal of Physiology 203, 775-781.
Jones, S. A., Adamson, S. L., Bishai, I., Lees, J., Engelberts, D. & Coceani, F. (1993). Eicosanoids in third ventricular cerebrospinal fluid of fetal and newborn sheep. American Journal of Physiology 264, R135-142
[Medline]
Joseph, S. A. & Walker, D. W. (1990). Catecholamine neurones in fetal brain: effects on breathing movements and electrocorticogram. Journal of Applied Physiology 69, 1903-1911
[Medline]
Joseph, S. A. & Walker, D. W. (1993). Effects of intracisternal monoamines on breathing movements in fetal sheep. American Journal of Physiology 264, R1139-1149
[Medline]
Kitterman, J. A., Liggins, G. C., Clements, J. A. & Tooley, W. H. (1979). Stimulation of breathing movements in fetal sheep by inhibitors of prostaglandin synthesis. Journal of Developmental Physiology 1, 453-466
[Medline]
Koos, B. (1985). Central stimulation of breathing movements in fetal lambs by prostaglandin synthetase inhibitors. The Journal of Physiology 362, 455-466
[Abstract]
Lumbers, E. R., Smith, F. G. & Stephens, A. D. (1985). Measurement of net transplacental transfer of fluid to the fetal sheep. The Journal of Physiology 363, 289-299.
Norton, J. L., Adamson, S. L., Bocking, A. D. & Han, V. K. M. (1996). Prostaglandin-H synthase-1 (PGHS-1) gene is expressed in specific neurons of the brain of the late gestation ovine fetus. Developmental Brain Research 95, 79-96.
[Medline]
Sebens, J. B. & Korf, J. (1975). Cyclic AMP in cerebrospinal fluid: accumulation following probenecid and biogenic amines. Experimental Neurology 46, 333-344
[Medline]
Smith, G. N., Brien, J. F., Homan, J., Carmichael, L. & Patrick, J. (1990). Indomethacin reversal of ethanol-induced suppression of ovine fetal breathing movements and relationship to prostaglandin E2. Journal of Developmental Physiology 14, 29-35
[Medline]
Stone, T. W. & Connick, J. H (1985). Quinolinic acid and other kynurenines in the central nervous system. Neuroscience 15, 597-617
[Medline]
Tamarkin, N. R., Goodwin, F. K. & Axelrod, J. (1970). Rapid elevation of biogenic amine metabolites in human CSF following probenecid. Life Science 9, 1397-1408.
Tze-Chan, T., MacLusky N. J. & Adamson, S. L. (1994). Ontogenesis of prostaglandin E2 binding sites in the brainstem of the sheep. Brain Research 652, 28-39
[Medline]
van Bel, F., Bartelds, B., Teital, D. F. & Rudolph, A. M. (1995). Effect of indomethacin on cerebral blood flow and oxygenation in the normal and ventilated fetal lamb. Pediatric Research 38, 243-250
[Abstract]
Wallen, L. D., Murai, D. T., Clyman, R. I., Lee, C. H., Mauray, F. E. & Kitterman, J. A. (1986). Regulation of breathing movements in fetal sheep by prostaglandin E2. Journal of Applied Physiology 60, 526-531
[Medline]
T. C. Tai and S. L. Adamson
Developmental changes in respiratory, febrile, and cardiovascular responses to PGE2 in newborn lambs
Am J Physiol Regulatory Integrative Comp Physiol,
June 1, 2000;
278(6):
R1460 - R1473.
[Abstract]
[Full Text]
[PDF]
This Article 
Abstract
Full Text (PDF)
Services
Email this article to a friend
Similar articles in this journal
Similar articles in PubMed
Alert me to new issues of the journal
Download to citation manager
Citing Articles
Citing Articles via HighWire
Citing Articles via Google Scholar
Google Scholar
Articles by Walker, D. W.
Articles by Pratt, N.
Search for Related Content
PubMed
PubMed Citation
Articles by Walker, D. W.
Articles by Pratt, N.
HOME
HELP
FEEDBACK
SUBSCRIPTIONS
ARCHIVE
SEARCH
TABLE OF CONTENTS