INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

PGE₂ induces diverse centrally mediated effects, including fever (3) and a sympathetically mediated increase in blood pressure and heart rate (HR; 13, 20, 42), in adults of many species. Interestingly, PGE₂ has prominent respiratory depressant effects in the perinatal period in humans, sheep, and pigs (15, 28, 30), whereas it appears to have little or no central respiratory effects in adults in that intracarotid PGE₂ has little effect on arterial PCO₂ (Pa_CO2) in adult sheep and dogs (20). Similarly, the centrally mediated febrile (3) and cardiovascular effects (5, 20) so prominent in adult animals are absent in the perinatal period (Refs. 19, 24, 37, and unpublished observations). The febrile response to PGE₂ emerges during the first 2 wk of postnatal life in the lamb (10). The early postnatal period may, therefore, be important in the transition from fetal to adult responses to PGE₂. This interval may be important because of the precipitous decrease in plasma and brain PGE₂ concentrations after birth in the sheep (23), which, in pigs, results in a postnatal upregulation of PGE₂ receptors and a shift in their subtype expression in brain synaptosomes (29).

In the current study we tested the hypothesis that transition from fetal to adult respiratory, cardiovascular, and febrile responses to intracarotid PGE₂ infusion occurs in the first 2 wk after birth in lambs. Second, we examined the mechanisms mediating ventilatory responses to intracarotid PGE₂ infusion in newborn lambs. More specifically, we tested the hypothesis that PGE₂ reduces central respiratory drive and central CO₂ sensitivity. We used intracarotid infusions of PGE₂ to elicit central responses, because this route is more effective than intracerebroventricular administration in eliciting central cardiovascular (5) and febrile (10) responses to PGE₂ in this species. In pathophysiological circumstances, the PGE₂ interacting with central PGE₂ receptors may enter the brain from blood, as in this experiment, or be synthesized locally in the brain (4, 10, 12, 22, 34, 44, 46, 51). Finally, to assist in differentiating peripheral from central effects, the responses to intrajugular PGE₂ infusion were also studied.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Animal surgery and experiments were approved by the Animal Care Committee of Mount Sinai Hospital and were conducted in accordance with guidelines approved by the Canadian Council on Animal Care.

Surgery and postoperative care. Twelve newborn lambs (Suffolk-Dorset crossbred) were anesthetized with halothane, intubated, and artificially ventilated 2 days after spontaneous vaginal delivery. With the use of sterile surgical techniques, a temperature probe (Physiotemp, Clifton, NJ) and a polyvinyl catheter were inserted via the left carotid artery. The catheter tip was advanced until the tip lay in the brachiocephalic artery. Correct placement was confirmed at necropsy. The carotid arterial catheter was used for intracarotid infusion to the cerebral circulation. Two catheters were inserted in the left jugular vein (1 for blood sampling and the other for intravenous infusion), and two catheters were inserted in the right femoral artery (1 for blood sampling and the other for blood pressure and HR monitoring). Stainless steel wire electrodes (Cooner Wire, Chatsworth, CA) were implanted in the thyroarytenoid (TA) muscles of the larynx to monitor activity of the main glottic adductor muscle. TA electrodes were placed bilaterally using a curved suture needle that pierced the thyroid cartilage at the ventral cricothyroid junction and exited just lateral to its ventral prominence as described previously (11). Electrodes were sewn into the diaphragmatic muscle through a small incision at the ninth intercostal space to monitor central respiratory output. After surgery, lambs were placed under a heat lamp and monitored closely until awake. A single dose of Temgesic (75 µg im; buprenorphine hydrochloride, Reckitt and Colman Pharmaceuticals, Hull, UK) was administered as a postoperative analgesic. Lambs received Pen-Di-Strep (0.5 ml im; 100,000 U procaine penicillin G, 125 g dihydrostreptomycin sulfate; Rogar STB, London, ON, Canada) twice daily for 1 wk as a prophylactic antibiotic. All catheters were filled with heparinized saline (40 U heparin sodium/ml 0.9% NaCl) and flushed daily to maintain patency. Lambs were bottle-fed three or four times daily with an artificial milk replacement (Frober, Cambridge, ON, Canada).

Experimental protocol. Experiments were performed 3 days after surgery when lambs were 5 days of age and were repeated again in the same animals at 10 and 15 days of age. Blindfolded lambs were studied suspended in a sling at an ambient temperature of 21 ± 1°C. They were not anesthetized or sedated. Variables were measured during four sequential 1-h periods. In the first hour, variables were recorded but not analyzed to allow time for the lambs to become acclimatized. During the second hour, saline was infused into the carotid arterial catheter. These data were used as control. During the third hour, PGE₂ was infused into the jugular venous catheter at 3 µg/min. During the fourth hour, PGE₂ was infused into the carotid arterial catheter at 3 µg/min. During the first 1/2 h of each period, uninterrupted cardiovascular and respiratory measurements were obtained. During the second 1/2 h of each period, tests were performed to measure metabolic rate, ventilatory response to central chemoreceptor (CO₂ rebreathing test), and peripheral chemoreceptor stimulation (KCN test) as described in detail below.

PGE₂ (Cayman Chemical, Ann Arbor, MI) was dissolved in double-distilled ethanol at 10 mg/ml and stored as a stock solution at 20°C. On the day of the experiment, 20 µl of stock PGE₂ was dried using prepurified liquid nitrogen (Liquid Carbonic, Toronto, ON, Canada) and reconstituted in 20 ml of 0.9% NaCl. PGE₂ or vehicle (saline) were infused at a rate of 0.3 ml/min using an infusion pump (model 944, Harvard Apparatus, Dover, MA).

Monitoring and data analysis. Respiratory and cardiovascular variables were recorded continuously on a polygraph chart recorder (model 7D, Grass Instruments). Signals from the polygraph were digitized (model 4000 Vetter, A. R. Vetter) and recorded on magnetic tape (VCR, Panasonic Hi-tech) for later analysis by computer. Arterial blood pressure was measured using a pressure transducer (model CDX3, Cobe, Lakewood, CO) calibrated with a mercury manometer. An analog filter was used to obtain mean arterial blood pressure (MABP). HR was derived from the pulsatile arterial pressure signal (tachograph model 7P4, Grass Instruments). Body temperature was recorded at 0, 15, 30, 45, and 55 min into each 1-h period from the indwelling arterial temperature probe.

Blood sampling. Femoral arterial blood samples (0.5 ml) were collected at 15, 30, and 55 min of each hour and measured immediately at 37°C in a blood gas analyzer (model 170, Corning Medical, Medfield, MA) and CO-oximeter (model 2500, Corning Medical).

Metabolic rate. A two-way Rudolph valve (model 2210; Hans Rudolph, Kansas City, MO) with a 6-liter Douglas bag attached to the expiratory outlet was connected to the end of the pneumotachograph. Expired gases were collected into the Douglas bag for a 2-min period, at which time the two-way valve was disconnected from the pneumotachograph. Mixed expired gases were analyzed by sampling from the Douglas bag and measured with a gas analyzer (model 170, Corning Medical). Total volume collected in the Douglas bag was measured by a spirometer. Oxygen consumption (O₂) and CO₂ production (CO₂) were calculated from expired gas tensions (PE_O2 and PE_CO2) (26). The respiratory quotient was the ratio of CO₂ to O₂.

Respiratory variables. Respiratory activity was measured by monitoring diaphragmatic electromyogram (dEMG) activity and airflow (AF). Thyroarytenoid EMG (TAEMG) activity was measured to assess upper airway (glottal) patency. TAEMG and dEMG were recorded using a Grass preamplifier (model 7P511), and the signals were then integrated (time constant 0.05 s; Grass integrator, model 7P3). These signals were analyzed by computer using a peak detection algorithm (Viewdac Sequences programmed by Danny Seto) to obtain peak integrated () TAEMG and dEMG. Respiratory AF was measured using a pneumotachograph (size 0, Fleisch, Lausanne, Switzerland) connected to a tightly fitting face mask. The flow signal from the pneumotachograph was measured with a volumetric pressure transducer (Grass, model PT5). With the use of the computer program, peak inspiratory AF was measured from this signal and it was also integrated (Grass integrator, model 7P10) to obtain continuous tidal volume (V_t) measurements from which breathing frequency (f), ventilation (VE = V_t × f), inspiratory time (TI = time from onset to cessation of AF) and expiratory time (TE = remainder of respiratory cycle) were determined. The dEMG signal was used to determine the number of breath-to-breath intervals of 2- to 10-s duration (i.e., short central apneas). Breath-to-breath intervals >10 s were rare. V_t, Pa_CO2, and mixed expired PCO₂ were used to calculate deadspace volume using the Bohr equation (26). End-tidal CO₂ (PET_CO2) was measured continuously by a CO₂ infrared analyzer (model IH31, San-Ei Instruments, Japan) that sampled gas (20 ml/min) from the outlet of the face mask.

Central chemoreceptor sensitivity. The ventilatory response to an increase in central chemoreceptor drive was examined by the Read's CO₂ rebreathing method (43). A rebreathing bag (250-300 ml) was initially filled with a gas mixture containing 7-10% CO₂ in O₂. The rebreathing bag was attached to the end of the pneumotachograph, and the lambs rebreathed from the bag for 2 min. Breath-by-breath response to the rise in alveolar CO₂ was analyzed by computer. Instantaneous minute ventilation (V_i) of each breath was calculated from the equation: V_i = (instantaneous respiratory frequency) × V_t. Breath-by-breath plot of V_i against PET_CO2 was obtained for each rebreathing test, and linear regression analysis of the relationship was used to calculate the correlation coefficient and the slope of the response (which reflects the CO₂ sensitivity). CO₂ threshold was calculated from the x-intercept of the plot (Fig. 1).

View larger version (19K): [in this window] [in a new window]   Fig. 1.   Example of ventilatory response to increased CO2 caused by rebreathing during saline () and PGE2 intracarotid infusions () obtained from 1 lamb (lamb O) at 5 days of age. m, Slope of CO2 response curve; r, correlation coefficient; B, apnea threshold (x-intercept); ETCO2, end-tidal CO2; Vi, instantaneous minute ventilation.

Peripheral chemoreceptor sensitivity. Peripheral chemoreceptors were stimulated by injecting a bolus of potassium cyanide (KCN) (6) into the jugular vein catheter (0.1 mg/kg; Fisher Chemicals, Fair Lawn, NJ). Mean E was calculated during the first 10 s of hyperventilation after the KCN bolus.

Statistical analysis. Respiratory and cardiovascular data were acquired on tape playback using the Viewdac Data Acquisition Program (Keithley Instruments, Tauton, MA) and analyzed by computer (Viewdac Sequences programmed by Danny Seto). Continuously measured respiratory and cardiovascular variables were averaged over the first 30 min of each infusion. Variables measured in arterial blood and measurements of body temperature were averaged over 1-h periods.

Data are expressed as means ± SE. Statistical significance of changes in each variable during saline, intrajugular, and intracarotid PGE₂ infusion at each age were determined by one-way, repeated-measures ANOVA followed by the Dunnett's multiple comparisons test with saline infusion as control. Developmental changes in control variables were analyzed by comparing data from saline control infusions at each age using a one-way, repeated-measures ANOVA followed by the Student-Newman-Keuls multiple comparisons test. Developmental changes in response to intrajugular and intracarotid PGE₂ infusion were determined by comparing the differences in infusion means from saline [i.e., (PGE₂ intrajugular  saline), (PGE₂ intracarotid  saline)] at each age using a one-way, repeated-measures ANOVA followed by the Student-Newman-Keuls multiple comparisons test. Developmental changes in the proportion of animals that exhibited periodic breathing and shivering during intracarotid PGE₂ infusion were analyzed by ² analysis. A t-test was used to compare variables under control conditions in lambs that later developed Biot periodic breathing vs. lambs that maintained regular breathing rhythms during intracarotid PGE₂ infusion. A P < 0.05 was considered significant.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Twelve lambs were studied at 5 days, nine lambs at 10 days, and ten lambs at 15 days after birth. All 12 lambs were not studied at each age because experiments were not performed if basal body temperature was >40°C to exclude potentially febrile animals from the study (10). The body weights of the lambs at the three ages are shown in Table 1.

During the saline control infusion, younger lambs (5 days) had a significantly higher ventilation per kilogram body weight and metabolic rate per kilogram than older lambs (10 and 15 days) (Table 1, Figs. 2 and 3). Younger lambs also had significantly higher Pa_CO2, bicarbonate, base excess, total hemoglobin and hematocrit, O₂ content, and significantly lower Pa_O2 than the older lambs (10 and 15 days) (Table 1). Low-amplitude phasic expiratory TAEMG activity was observed during saline infusion at all ages (e.g., see Figs. 6 and 8). The mean peak amplitude of TAEMG activity during saline infusion did not change significantly with age (Fig. 2). Other measured variables also showed no significant changes between 5 and 15 days.

View larger version (39K): [in this window] [in a new window]   Fig. 2.   Effects of PGE2 on respiratory variables. A: respiratory variables during saline infusion (open bars) and PGE2 infused to jugular vein (shaded bars) and carotid artery (solid bars) at each age [5-, 10-, and 15-day (d)-old lambs]. * Statistically significant change from control (saline infusion) within each age group. Significant age-related changes in control variables are indicated by open bars that have different lowercase letters. B and C: changes in response to intrajugular and intracarotid PGE2 infusion with age, expressed as a difference from control [i.e., (PGE2 intrajugular  saline) and (PGE2 intracarotid  saline)]. Values are means ± SE. E, minute ventilation; TAEMG, peak integrated thyroarytenoid electromyographic activity; PaCO2 and PaO2, arterial PCO2 and PO2, respectively.

View larger version (31K): [in this window] [in a new window]   Fig. 3.   Effects of PGE2 on metabolism. A: variables during saline infusion (open bars) and PGE2 infused into jugular vein (shaded bars) and carotid artery (solid bars) at each age (5-, 10-, and 15-day-old lambs). * Statistically significant change from control (saline infusion) within each age group. Significant age-related changes in control variables are indicated by open bars that have different lowercase letters. B and C: changes in response to intrajugular and intracarotid PGE2 infusion with age, expressed as a difference from control [i.e., (PGE2 intrajugular  saline) and (PGE2 intracarotid  saline)]. Values are means ± SE. Temp, body temperature; O2, oxygen consumption; CO2, CO2 production.

Intracarotid PGE₂ had many respiratory, thermoregulatory, and cardiovascular effects, which are described in detail below. In addition, intracarotid PGE₂ had a sedative effect in all lambs as well as caused diarrhea and salivation in some lambs at all ages; these effects were not observed during intrajugular PGE₂ infusion.

Respiration. Intrajugular PGE₂ infusion did not significantly alter ventilation, breathing frequency, V_t, the mean peak amplitude of dEMG activity, or the number of short apneic pauses (2 to 10 s in duration) from the control values shown in Table 1. However, at all ages, intrajugular PGE₂ slightly increased PET_CO2, Pa_CO2 (+1.3 ± 0.6 mmHg), and the mean peak amplitude of TAEMG activity and slightly decreased Pa_O2 (5.0 ± 2.1 mmHg), although changes were significant only at 5 and/or 10 days (Fig. 2). All other arterial variables (pH, O₂ saturation, O₂ content, HCO₃, base excess, hematocrit, and total hemoglobin) remained unchanged from control values (Table 1) during intrajugular PGE₂.

Intracarotid PGE₂ caused marked and immediate changes in respiration in contrast to the minor effects observed during intrajugular infusion. The onset of carotid arterial infusion caused a brief central apnea in most lambs (Fig. 4). Breathing resumed with a markedly abnormal AF pattern (Fig. 4) that was likely the result of the sudden and marked increase in the peak amplitude of the phasic TAEMG activity that also occurred at this time (Fig. 5). Most of the increase in TAEMG activity occurred during early expiration when alterations in AF were most marked (Fig. 6). Increased TAEMG activity and AF disruptions, however, were sometimes also observed during inspiration (Fig. 7). The mean peak amplitude of TAEMG activity increased significantly by more than twofold when averaged over the first 30 min of intracarotid PGE₂ infusion at all ages compared with control (Fig. 2).

View larger version (40K): [in this window] [in a new window]   Fig. 4.   Polygraph tracings showing airflow from all 12 lambs (A-O) at 5 days during intracarotid saline infusion (tracings on left) and during the last 15 s of intrajugular PGE2, followed by first 75 s of intracarotid PGE2 infusion (tracings on right). Onset of intracarotid PGE2 usually caused a brief apnea, followed by irregular breathing and interruptions in expiratory airflow. Inspiratory airflow is upward. Ticks on time axis are 1 s apart.

View larger version (44K): [in this window] [in a new window]   Fig. 5.   Polygraph tracings showing immediate changes in airflow (top), integrated TAEMG (TA) activity (middle), and integrated diaphragm EMG (dEMG) activity (bottom) after starting infusion of intracarotid PGE2 (indicated by arrowhead on time axis) in 1 lamb (lamb I) at 5 days of age. Intracarotid PGE2 caused an immediate increase in TAEMG activity and expiratory airflow braking in all lambs studied. Inspiratory airflow is upward. Ticks on time axis are 1 s apart.

View larger version (37K): [in this window] [in a new window]   Fig. 6.   Playback of digitized data showing dEMG, TAEMG, and airflow during intracarotid saline (A) and intracarotid PGE2 infusion (B) in 1 lamb at 5 days of age. Note marked disruption in airflow during expiratory phase of respiration that was associated with a marked increase in TA muscle activity during intracarotid PGE2 (B). Pattern of airflow during inspiration is also altered in association with an increase in TAEMG activity during intracarotid PGE2 infusion, but effect is less marked than during expiration. Inspiratory airflow is upward.

View larger version (67K): [in this window] [in a new window]   Fig. 7.   Relationship between dEMG (Int dia EMG), TAEMG (Int. TAEMG), and airflow (positive flow is inspiratory, negative flow is expiratory) during Biot periodic breathing is shown on an expanded time scale (lamb E at 5 days of age). As shown in this example, intracarotid PGE2 sometimes caused increased TAEMG activity and associated airflow disruptions during both early inspiration and early expiration.

Intracarotid PGE₂ also caused marked disruptions in central respiratory rhythm. The number of short apneic pauses (2 to 10 s in duration) during the first 30 min of infusion significantly increased at all ages from control (Table 1) to 36 ± 14 at 5 days, 41 ± 16 at 10 days, and 48 ± 23 at 15 days (no significant change with age). Also, a Biot periodic breathing pattern began at some point during the 1-h infusion in approximately half the lambs (7 of 12 lambs at 5 days, 4 of 9 at 10 days, and 4 of 10 at 15 days; no significant change with age) (Fig. 8 and see Fig. 7). Biot periodic breathing is a cluster-type periodic breathing characterized by groups of breaths that start and stop abruptly (52). TAEMG activity during apneas in lambs exhibiting a Biot periodic breathing pattern was either sustained or was very high at the start of the apnea and then decreased with time (e.g., Fig. 8B) in three of seven lambs at 5 days of age and two of four lambs at 15 days of age. In other cases, TAEMG activity was markedly reduced or absent during apneas [1 at 5 days, 1 at 15 days (different lambs); e.g., Fig. 8A] or no consistent pattern was observed (3 at 5 days, 1 at 15 days). The time of onset, duration of Biot periodic breathing, and number of breaths per cluster was variable between animals. Interestingly, in these lambs, hypercapnia induced by CO₂ rebreathing did not restore continuous breathing (e.g., Fig. 9), whereas breathing was always briefly restored after KCN injection (not shown).

View larger version (12K): [in this window] [in a new window]   Fig. 8.   Polygraph tracings of TAEMG and dEMG during intracarotid saline control and intracarotid PGE2 infusion in same lamb (lamb E) studied at 5 days of age (A) and at 15 days of age (B). Each segment is 1 min in duration. Scale bars show length of a 200-µV calibration signal. Intracarotid PGE2 infusion induced Biot periodic breathing in ~50% of lambs studied at each age. In some studies, TAEMG activity was consistently low during apneas as in A; in others, TAEMG was consistently high during apneas as in B; in others, no consistent relationship was observed. During saline infusion, a low-amplitude phasic pattern of TAEMG activity was observed [more prominent in B because of increased amplification of signal (note scale bars)] as well as some large peaks that were likely associated with swallowing. During infusion of intracarotid PGE2, there was little change in peak amplitude of dEMG activity, whereas the amplitude of phasic changes in TAEMG during breathing was consistently increased.

View larger version (31K): [in this window] [in a new window]   Fig. 9.   Example tracings obtained during CO2 rebreathing tests during intracarotid saline (A) and intracarotid PGE2 (B) from a 5-day-old lamb (lamb M). During intracarotid saline, steady increase in end-tidal CO2 during rebreathing test was associated with a steady increase in both breathing frequency and amplitude of dEMG. During intracarotid PGE2, apneas often persisted despite hypercapnia induced by rebreathing as shown here.

Intracarotid PGE₂ significantly decreased ventilation (~18%) during the first 30 min of infusion at 5 days, whereas ventilation did not change at 10 days and tended to increase at 15 days (Fig. 2). The developmental attenuation in the hypoventilatory response to intracarotid PGE₂ was significant (Fig. 2). Ventilation decreased at 5 days due to a significant decrease in breathing frequency (from 68 ± 4 breaths/min during saline to 53 ± 2 breaths/min during intracarotid PGE₂). Tidal volume and peak dEMG activity did not change (not shown). The decrease in breathing frequency appeared to be due to a nonsignificant 40% increase in TE in that there was almost no change in the mean TI (not shown). No significant changes in breathing frequency, V_t, and peak dEMG activity were observed during intracarotid PGE₂ at 10 and 15 days.

At all three ages, intracarotid PGE₂ caused significant hypercapnia (+10 ± 2 mmHg) and hypoxia (13 ± 2 mmHg) (Fig. 2) and acidosis (0.12 ± 0.05 pH unit) as well as significant decreases in O₂ saturation (6 ± 1%), O₂ content (1.3 ± 0.2 ml/dl), and total hemoglobin concentration (0.3 ± 0.1 g/dl) from control values (Table 1) during the 1-h period of intracarotid infusion. Values in parentheses in this paragraph show results for the three ages combined. Only the change in Pa_CO2 changed significantly with age; the size of the increase progressively diminished from 5 to 15 days [(PGE₂ intrajugular  saline) in Fig. 2].

Chemosensitivity. Intrajugular PGE₂ reduced central CO₂ sensitivity by 20% in lambs at 5 days of age, whereas no significant change was observed in older lambs. Ventilation during KCN stimulation was not significantly different from control (Table 1) during intrajugular PGE₂ infusion at any age.

Intracarotid PGE₂ decreased CO₂ sensitivity, and this effect significantly diminished with increasing age; CO₂ sensitivity decreased significantly by 70% in 5-day-old lambs, 27% in 10-day-old lambs and nonsignificantly by 13% in 15-day-old lambs (Fig. 10). The correlation between V_i and PET_CO2 was significantly weaker during intracarotid PGE₂ at all ages, and, as for CO₂ sensitivity, this effect waned significantly between 5 and 10 days of age (Fig. 10). The threshold of the central CO₂ response increased significantly by ~10 mmHg during intracarotid PGE₂ infusion at all ages, but this effect did not change significantly with age (Fig. 10).

View larger version (37K): [in this window] [in a new window]   Fig. 10.   Effect of PGE2 on central chemoreceptor sensitivity as determined by CO2 rebreathing test, showing CO2 sensitivity (top), correlation coefficient (middle), and threshold (bottom). A: variables during saline infusion (open bar) and PGE2 infused into jugular vein (shaded bar) and carotid artery (solid bar) at each age (5-, 10-, and 15-day-old lambs). * Statistically significant change from control (saline infusion) within each age group. Significant age-related changes in control variables are indicated by open bars that have different lowercase letters. B and C: changes in response to intrajugular and intracarotid PGE2 infusion with age, expressed as a difference from control [i.e., (PGE2 intrajugular  saline) and (PGE2 intracarotid  saline)].

Although intracarotid PGE₂ infusion decreased ventilation by 18% and reduced CO₂ sensitivity by 70% in 5-day-old lambs, ventilation during peripheral chemoreceptor stimulation with KCN was unchanged from control (Table 1) at all ages. However, because ventilation was reduced during intracarotid PGE₂ only in 5-day-old lambs, whereas ventilation after KCN was not, this means that the increase in ventilation after KCN during intracarotid PGE₂ was relatively greater in 5-day-old lambs than in older lambs.

Metabolism and thermoregulation. Intrajugular PGE₂ did not change O₂ or CO₂ from control values (Table 1, Fig. 3) in 5- and 10-day-old lambs, whereas marked increases in both variables were observed at 15 days. The increase in CO₂ at 15 days was significantly greater than in the younger lambs (Fig. 3). The respiratory quotient did not change from control values (Table 1) at any age. The incidence of shivering significantly increased with age; mild shivering was observed in 2 of 12 lambs at 5 days, 4 of 9 lambs at 10 days, and vigorous shivering was observed in all 10 lambs studied at 15 days.

Intracarotid PGE₂ did not significantly alter O₂ or CO₂ from control values (Table 1, Fig. 3) in 5- and 10-day-old lambs. However, in 15-day-old lambs, CO₂ production was significantly elevated during intracarotid PGE₂ infusion. Also, there was a small but significant increase in the effect of PGE₂ on O₂ with increasing age (Fig. 3). A similar trend in CO₂ was not statistically significant (Fig. 3). The respiratory quotient did not change from control values (Table 1) at any age.

In 5-day-old lambs, the fever that developed during intrajugular PGE₂ infusion was not maintained during intracarotid PGE₂ infusion (Fig. 3). In contrast, body temperature remained significantly elevated during intracarotid PGE₂ infusion in 10- and 15-day-old lambs (Fig. 3). As observed during intrajugular PGE₂ infusion, during intra-arterial infusion the proportion of lambs that exhibited shivering significantly increased with age. Mild shivering was observed in 4 of 12 lambs at 5 days, 5 of 9 lambs at 10 days, and vigorous shivering was observed in all 10 lambs at 15 days.

Cardiovascular responses. Intrajugular PGE₂ infusion caused a small but significant decrease in MABP (2.2 ± 0.7 mmHg) and a significant rise in HR (26 ± 6 beats/min) in 5-day-old lambs relative to control values (Table 1). Similar changes in MABP and HR were also observed at 10 and 15 days, but were not statistically significant (not shown).

During intracarotid PGE₂ infusion, there were small, nonsignificant increases (3 ± 3 mmHg) in MABP at all ages relative to control values (Table 1). HR remained near control values (Table 1) at 5 days; however, there was a trend toward an increase in HR (20 ± 17 beats/min) at 10 and 15 days so that the tachycardic response to intracarotid PGE₂ increased significantly with age between 5 and 10 days of age.

Biot vs. non-Biot breathing animals. Intracarotid PGE₂ induced Biot periodic breathing in ~50% of lambs studied at 5, 10, and 15 days, but this response was not consistently observed in the same lambs at different ages. We wondered whether parameters under control conditions were associated with the later development of Biot periodic breathing during intracarotid PGE₂ infusion. We found that at 5 days, lambs that later developed Biot periodic breathing had significantly lower control Pa_CO2 (45 ± 2 vs. 51 ± 2 mmHg) and PET_CO2, higher pH (7.46 ± 0.02 vs. 7.42 ± 0.01), and higher breathing frequency (77 ± 4 vs. 57 ± 3 breaths/min) and lower TE than lambs that maintained a regular breathing rhythm during intracarotid PGE₂. However, at 10 and 15 days, lambs that later developed Biot periodic breathing had significantly higher control Pa_O2 (84 ± 2 vs. 77 ± 2 mmHg at 10 days and 84 ± 2 and 76 ± 2 mmHg at 15 days) than lambs that maintained a regular breathing rhythm. There were no other significant differences between Biot breathing and regular breathing lambs at control for all other cardiovascular, metabolic, and respiratory variables measured in this study.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

This study shows marked developmental changes in respiratory and febrile responses to intracarotid infusions of PGE₂ in lambs between 5 and 15 days after birth. Intracarotid PGE₂ reduced central CO₂ sensitivity, decreased breathing frequency, induced hypercapnia, and induced a poorly sustained febrile response at 5 days, whereas by 15 days, these respiratory effects had waned and the fever was sustained. In contrast, central cardiovascular responses were slight throughout this age range, although there appeared to be a trend toward the development of a tachycardic response in the older lambs similar to adult sheep (5). This study was also the first to demonstrate that intracarotid PGE₂ causes an immediate, marked increase in TAEMG activity, impeding AF during expiration and sometimes inspiration as well, and that it can induce a Biot periodic breathing pattern in conscious lambs throughout the age range studied. Interestingly, although the capacity of PGE₂ to reduce central CO₂ sensitivity waned rapidly with postnatal age, the capacity of PGE₂ to disrupt breathing rhythm and impede AF was sustained for at least 2 wk postnatally.

In the current study, PGE₂ was infused peripherally. However, in vivo, the PGE₂ that interacts with PGE₂ receptors in the brain may arise from either a central or peripheral source. The cyclooxygenase enzymes required for PG production are localized within specific brain nuclei in fetal and adult sheep (4, 34), and the fetal and adult brain produce PGE₂ in vivo (10, 51) and in vitro (46), indicating that PGE₂ synthetase is also present. There is a striking overlap between the localization of putative PGE₂ receptors (50) and cyclooxygenase immunoreactivity (34) in specific brain regions of the perinatal brain, including those responsible for respiratory control (50). Thus locally produced PGE₂ may serve a neuromodulatory role within specific brain regions both before and after birth. Locally produced PGE₂ may also have important cerebral vasodilatory effects, such as during artificial ventilation at high airway pressure, severe hypotension, or combined hypercapnia and hypoxia (8, 27, 32). During each of these interventions in newborn pigs, PGE₂ increased three to five times in cerebrospinal fluid (CSF) or sagittal vein plasma. Peripherally produced PGE₂ may also interact with central receptors. Peripheral PGE₂ mediates central effects (3, 5, 10, 20, 21, 25) and influences CSF PGE₂ concentrations (10), although entry is impeded by the blood-brain/blood-CSF barrier (22).

We used peripheral infusions to stimulate PGE₂ receptors to compare responses elicited at different ages. Although artificial in the current study, high circulating levels of PGE₂ are normal in late fetal life (fetal levels are ~10 times higher than in lambs and even higher during labor) (23, 45) and in adults after exposure to pyrogens (levels increase ~4 times) (12). Therefore, in the current study, the 40 times increase in plasma PGE₂ concentration, predicted to have occurred on the basis of prior work in lambs (15), may have been pharmacological. However, these plasma levels may have been necessary to achieve pathophysiological levels in brain, because entry from plasma is impeded by the blood-brain/blood-CSF barrier and because in pathophysiological situations (e.g., fever), brain levels are likely augmented by local PGE₂ synthesis. Indeed, the 40 times increase in plasma concentration would be predicted to increase CSF PGE₂ concentration 10 times (22), which is less than the 20 times increase in CSF PGE₂ that occurred during intravenous endotoxin exposure in lambs (10).

Respiration. Intrajugular PGE₂ significantly affected some respiratory variables, but in all cases responses to intrajugular PGE₂ were similar but of smaller magnitude than those induced by intra-arterial PGE₂. Thus it is likely that these effects were elicited by the small amount of PGE₂ that escaped pulmonary inactivation (36), thereby entering the arterial circulation and ultimately reaching the brain. If so, the brain regions mediating these effects are highly sensitive and/or highly accessible to PGE₂ in arterial blood because ~90% of the infused PGE₂ would have been catabolized by the lung (36). An action of PGE₂ on the brain is also supported by the persistence of the inhibitory effects of PGE₂ on breathing in fetal sheep after peripheral chemoreceptor denervation, vagotomy, and brain stem transection at the level of the upper pons/inferior colliculus (21, 25), by the high density of putative PGE₂ receptors localized within respiratory regions of the perinatal sheep brain stem (50), and by the inhibition of fetal breathing caused by a selective increase in brain PGE₂ concentrations (51). In addition, in isolated brain stem-spinal cord preparations from newborn rats, PGE₁ inhibits inspiratory nerve activity of the ventral respiratory group neurons (1), although this activity may be more closely related to in vivo gasping than eupneic respiration (47).

Intracarotid PGE₂ in 5-day-old lambs caused hypoventilation due to a decrease in breathing frequency. It also caused hypoxemia, hypercapnia, and increased the incidence of short apneas (2-10 s). Similar findings were observed during intra-aortic PGE₂ infusion in awake newborn lambs (2-6 days old) (15), except that hypoventilation in this slightly younger age group was more marked and was caused by reductions in V_t as well as breathing frequency. The current study shows that this age-related attenuation in respiratory inhibition continues so that by 10 days after birth, PGE₂ no longer caused hypoventilation, and by 15 days ventilation tended to increase. Interestingly, although the hypercapnic response to intracarotid PGE₂ infusion attenuated with age, it was still significant at 15 days when ventilation was no longer affected. This cannot be entirely explained by a PGE₂-induced increase in metabolic rate, so an increase in deadspace ventilation or a ventilation-perfusion mismatch may be involved. Alternatively, due to the shape of the ventilation hyperbola, an immeasurable change in ventilation may have resulted in a measurable change in PCO₂. In adults, intravenous PGE₂ increases ventilation (31), whereas intracarotid PGE₂ appears to have little effect [i.e., there is little change in Pa_CO2 (20)]. The waning of the hypoventilatory response to PGE₂ was likely not due to the increase in body weight with development (hence greater volume dilution for infused PGE₂), because the change in body weight was relatively small (+30%) and because the ability of intracarotid PGE₂ to induce Biot periodic breathing, short apneas, and to enhance TAEMG activity did not change with age.

Sedation and/or anesthesia have been shown previously to induce tonic TAEMG during apneas, reduce central respiratory drive, and reduce central CO₂ sensitivity (18, 35, 41). Because sedation was observed during intracarotid PGE₂ infusion, sedation may have contributed to these responses. However, the degree of sedation and the incidence of tonic TAEMG activity during apnea were not noticeably altered with age, whereas "respiratory depression" was most pronounced at 5 days and had waned considerably by 15 days. Thus it is unlikely that the sedative effects of PGE₂ can entirely account for the respiratory responses observed in the current study.

Chemoreceptor sensitivity. Intracarotid PGE₂ caused Pa_CO2 to increase at all ages; at 5 days the increase was due to a reduction in ventilation with constant metabolic rate, whereas at 15 days, the increase was due to an increase in metabolic rate with a weaker increase in ventilation. At both ages, the mismatch between ventilation and metabolism was associated with a decrease in slope and/or an increase in threshold of the central chemoreceptor response to CO₂, whereas the ventilatory response to peripheral chemoreceptor stimulation was not attenuated at any age. These results are in agreement with a preliminary report (49) that showed that PGE₂ decreased the ventilatory response to 5% CO₂ (i.e., central chemoreceptor sensitivity) but did not alter the ventilatory response to 10 min of hypoxia (i.e., peripheral chemoreceptor sensitivity) in 3-day-old sedated newborn piglets. Possibly PGE₂ reduces central CO₂ sensitivity by increasing cerebral blood flow and, hence, increasing CO₂ clearance from the brain. However, at least in piglets, this mechanism seems unlikely, because, although PGE₂ increases cerebral blood flow by ~20%, the increase in flow is accompanied by an equivalent increase in cerebral metabolic rate (9) so that normal CO₂ clearance is maintained. Given the importance of the pneumotaxic center in the pons in integrating chemoreceptor afferent activity (33, 48), it is more likely that the inhibitory effects of PGE₂ on central chemoreceptor sensitivity are mediated by putative PGE₂ receptors that are localized in high density near the pneumotaxic center in newborn lambs (50).

Biot periodic breathing. The current study provides the first evidence that a biologically active substance that is produced by the brain is capable of inducing Biot periodic breathing in conscious experimental animals. The pneumotaxic center in the pons is likely involved because there is a high density of putative PGE₂ receptors localized near this site (50) and because Biot periodic breathing is induced in approximately one-third of adult cats after ablating this region under deep pentobarbital sodium anesthesia (52). The inability to induce Biot periodic breathing in all pneumotaxic-ablated cats was thought to be due to the complex interaction of vagal afferents and chemoreflexes (52). Intracarotid PGE₂ also induced Biot periodic breathing in only a fraction of lambs studied (~50%). Susceptible lambs were relatively hypocapnic (5 days) or hyperoxic (10 and 15 days) under control conditions, suggesting that basal respiratory activity in these lambs was augmented. Vagal, trigeminal, or arousal mechanisms may have been responsible because control peripheral and central chemosensitivities were not different. The mechanism whereby susceptibility to PGE₂-induced Biot periodic breathing was increased in these lambs is not known. Stimulation of the peripheral chemoreceptors, but not the central chemoreceptors, was able to reverse PGE₂-induced Biot periodic breathing in lambs, as described previously in pneumotaxic-ablated cats (52). The similarity in respiratory effects of pneumotaxic ablation and PGE₂ infusion suggests that PGE₂ may induce Biot periodic breathing by disruption of afferent or efferent signal processing within the pneumotaxic center. We speculate that endogenously produced PGE₂ acting at this site disrupts normal respiratory rhythmogenesis in septic and/or premature newborn infants. Indeed, this may be the mechanism whereby a PG synthesis inhibitor promoted regular breathing in a human infant with apnea of prematurity (16).

TAEMG activity. This study also shows for the first time that PGE₂ enhances expiratory TAEMG activity, thereby resulting in AF braking during expiration. Expiratory TAEMG activity in awake lambs is tonically inhibited by vagal afferents (39), but this activity can also be enhanced by vagal afferents in response to pulmonary deflation (17) or pulmonary edema (40). In the current study, the effect on TAEMG activity was greater during intracarotid than intrajugular infusion, and the augmentation of activity occurred within seconds after starting intracarotid infusion and thus before arterial blood gas tensions could change or significant edema could develop. In adults, stimulation of pulmonary irritant, C fibers (formerly J receptors), and/or stretch receptors can also alter expiratory laryngeal activity; however, such reflexes usually elicit marked cardiovascular and/or tachypneic responses (2), and these were not observed in the current study. Therefore, it seems unlikely that stimulation of vagal afferents mediated the increase in TAEMG activity. Although a peripheral mechanism cannot be ruled out, it is likely that TAEMG activity was augmented primarily by a central mechanism. Possibly putative PGE₂ receptors near the nucleus ambiguus (50) directly influenced activity in the motoneurons supplying the laryngeal muscles (7) or receptors localized near the nucleus of the solitary tract (NTS; 50) indirectly influenced activity by altering the processing of vagal afferent information (7).

Metabolism and thermoregulation. PGE₂ induced fever at all ages; however, there did appear to be age-dependent changes. In 5-day-old lambs, the febrile response elicited by intrajugular PGE₂ was not sustained during intracarotid PGE₂, and, in 5- and 10-day-old lambs, the rise in body temperature appeared to be due entirely to reduced heat loss, because metabolic rate was unaltered. In contrast, older lambs (15 days) shivered more vigorously and increased their metabolic rate so that an increase in heat production likely played an important role in generating their fever. In the youngest lambs, body temperature may have normalized in the final hour of the study, because the vasomotor responses that reduced heat loss could no longer be sustained (i.e., a time effect) or because the intracarotid route was less effective than the intrajugular route in eliciting fever. A definitive conclusion is not possible because of the nonrandomized study design. If the jugular route was more effective, then fever at this age may not have been mediated directly by an effect of PGE₂ on the brain but rather by a peripheral mechanism as proposed previously in lambs (38). Possibly, fever was elicited by stimulation of peripheral vagal afferents, an effect that would be more pronounced during intrajugular infusion. A vagal mechanism has been proposed to explain the rapidity of fever onset in adult guinea pigs (3). Whether a vagal mechanism is involved in fever in sheep is not known.

Cardiovascular responses. This study reports for the first time that PGE₂ does not elicit a central cardiovascular response in newborn lambs. A central cardiovascular response is also absent in fetal sheep (19, 24). PGE₂ mediates a central tachycardic and pressor response in adults of a number of species (5, 20). In adult sheep, intracarotid infusion of PGE₂ at lower doses (10-100 ng · min^-1 · kg^-1) than those used in the current study (~700 ng · min^-1 · kg^-1) caused significant elevations in both arterial blood pressure (25-35%) and HR (10-30%) (5, 20). In contrast, intracarotid PGE₂ infusion in newborn lambs (5-15 days old) caused small, but insignificant, increases in both blood pressure (<4%) and HR (<6%). Although there was a significant increase in the HR response with increasing age in the current study, most of the developmental maturation of the central tachycardic and pressor response to PGE₂ appears to arise after 2 wk postnatally in sheep.

Both the cardiovascular and febrile effects of PGE₂ are likely initiated by higher brain regions (10, 13). Nevertheless, the NTS in the brain stem serves as a major center involved in integration of visceral afferent information with efferent metabolic and cardiorespiratory information (3, 7, 14). PGE₂ receptors localized within the NTS (50) may therefore be involved in modulating respiratory, cardiovascular, and febrile responses initiated by PGE₂ acting at higher brain sites. In adult rats, bilateral lesions to the NTS attenuate both the central tachycardic and febrile responses induced by intracerebroventricular infusion of PGE₂, presumably by interrupting the neuronal pathways involved in the genesis of fever (14). Thus PGE₂ receptors and subtypes present within the NTS may modulate cardiovascular and febrile efferent output from higher brain regions.

In summary, the current study shows that during the early neonatal period in lambs there is a gradual transition from fetal to adult centrally mediated responses to PGE₂. Specifically, results show that PGE₂ causes hypoventilation by reducing breathing frequency and reduces central CO₂ sensitivity in lambs 5 days after birth and that these responses wane by 15 days. However, the PGE₂-induced increase in TAEMG activity, number of short apneas, and the incidence of Biot periodic breathing did not change with age over this interval. Although PGE₂ caused fever at all ages studied, fever was associated with an increase in metabolic rate, hence increased heat production, at 15 days but not at 5 days. The central tachycardic and pressor responses observed in adults appear to develop later in postnatal life. Currently, the mechanisms involved in the developmental changes in responses to PGE₂ are not known, but we speculate that it may involve developmental changes in central expression of PGE₂ receptor subtypes triggered by the precipitous decrease in brain PGE₂ concentrations in the early postnatal period.

Perspectives