CO2-induced expression of c-fos in the nucleus of the solitary tract and the area postrema of developing swine

Brain Research 837 Ž1999. 106–116 www.elsevier.comrlocaterbres

Research report

CO 2-induced expression of c-fos in the nucleus of the solitary tract and the area postrema of developing swine Anthony L. Sica a

a, )

, Phyllis M. Gootman b , David A. Ruggiero

c,d

Department of Medicine, Pulmonary and Critical Care DiÕision, Long Island Jewish Medical Center, Long Island Campus of the Albert Einstein College of Medicine, New Hyde Park, NY 11040 USA b Department of Physiology and Pharmacology, State UniÕersity of New York Health Science Center at Brooklyn, Brooklyn, NY 11203 USA c Departments of Psychiatry and Anatomy and Cell Biology, Columbia UniÕersity College of Physicians and Surgeons, New York, NY 10032 USA d Neurological Research Institute of Lubec, Lubec, ME 04652 USA Accepted 18 May 1999

Abstract This investigation was performed to determine whether hypercapnic exposure elicited expression of the c-fos protooncogene product, FOS, in nucleus of the solitary tract ŽNTS. and area postrema ŽAP. neurons of developing swine. Mean arterial blood pressure ŽMAP. and heart rate ŽHR. were also monitored to evaluate whether numbers of neurons containing FOS were related to changes of MAP and HR. In each experiment, two litter-matched piglets were prepared simultaneously, i.e., Saffan anesthesia, paralysis, and artificial ventilation Ž100% O 2 .. One animal was exposed to hypercapnia Ž1 h of 10% CO 2 , balance oxygen., while the other continued to breathe 100% O 2 . Animals were studied at three different ages: 5–8 days, 13–15 days, and 26–34 days old. In the NTS, FOS expression was prominent in regions corresponding to the general visceral afferent subdivision; the AP showed no such topographic distribution. The number of NTS and AP neurons with FOS in hypercapnic-exposed animals was significantly greater than those of unexposed animals. However, an age-related increase of FOS was observed only for NTS neurons, with the greatest number observed in 13- to 15-day-old animals. Increases of MAP, not HR, were noted during the early part of hypercapnia in the 5- to 8-day-old group; older animals exhibited no change of MAP. Our findings demonstrated that prolonged hypercapnic stimulation elicited FOS expression in AP and NTS neurons of developing animals, and that such expression was non-uniform, depending upon the region studied. q 1999 Published by Elsevier Science B.V. All rights reserved. Keywords: Development; Piglet; Hypercapnia; c-fos; Nucleus of the solitary tract; Area postrema

1. Introduction The maintenance of cardiovascular homeostasis is dependent to a large degree upon regulatory neural networks of the brainstem. These networks are able to maintain a relatively stable physiological state by mounting rapid responses to exogenous andror endogenous stressors. For example, increased levels of inspired CO 2 have been shown to activate brainstem neurons, including those that may act as central chemoreceptors for CO 2 w1,8,23,28,29x. The nucleus of the solitary tract ŽNTS. is a region harboring neurons that respond to such stimulation via postsynaptic activation as well as neurons subserving a putative chemosensory transducer function w8x. Such a finding ) Corresponding [email protected]

author.

Fax:

q 1-718-470-1035;

E-mail:

adds to the importance of the NTS as an integrative area, i.e., it receives numerous other afferent inputs, and via, oligo- and monosynaptic relations with other structures, participates in cardiovascular and respiratory regulation w6,9,11,24,33x. The area postrema ŽAP. is one of the regions that shares synaptic relationships with the NTS and is important for homeostatic regulation. While the AP has long been recognized as a chemoreceptive trigger zone for emesis w3x, recent physiological experiments showed that the AP may also be involved in respiratory and cardiovascular regulation, especially under stressful conditions w2,16,31x. In fact, both the NTS and AP have been implicated in cardiorespiratory regulation during conditions of increased chemical drive by anatomical studies using expression of the c-fos protooncogene product FOS as a marker of neuronal activation w13x. However, the degree of involvement is dependent upon the type of stimulation,

0006-8993r99r$ - see front matter q 1999 Published by Elsevier Science B.V. All rights reserved. PII: S 0 0 0 6 - 8 9 9 3 Ž 9 9 . 0 1 6 4 0 - 6

A.L. Sica et al.r Brain Research 837 (1999) 106–116

107

i.e., hypoxia or hypercapnia. While both NTS and AP neurons expressed FOS following prolonged hypoxia, only NTS neurons reliably expressed FOS after prolonged hypercapnia w19,35,40x. The number of AP neurons expressing FOS in response to hypercapnia was either very modest or absent, however, this finding may be due in part to the use of different species as experimental models w19,35,40x. While experiments using FOS as an index of activation have provided much information about brainstem structures responsive to hypercapnia in mature animals, little has been done to elucidate the roles of the AP and NTS in the response of immature animals. Such information is important because immature animals are much more likely than mature animals to encounter environmental conditions associated with increased levels of inspired CO 2 , e.g., sleep position, head covering, proximity to mother’s expired breath, which may initiate agonal events associated with sudden infant death w7,25,27,30x. Thus, in the present investigation, we carried out experiments to determine whether both NTS and AP neurons of developing swine expressed FOS in response to prolonged hypercapnia, and whether such expression showed any relationship to age. To evaluate whether degree of FOS expression was related to changes in peripheral measures of cardiovascular activity during hypercapnia, mean arterial pressure ŽMAP. and heart rate ŽHR. were measured in each experiment. Such a study has not been performed heretofore, and would yield data leading to a better understanding of NTS and AP contributions to the development of reflexive responses to chemoreceptor activation.

ratory Animal Resources of the Health Science Center at Brooklyn, State University of New York. Animals were assigned to one of three age groups: Ža. age group 1, 5–8 days old Ž n s 8.; Žb. age group 2, 13–15 days old Ž n s 8.; Žc. age group 3, 26–34 days old Ž n s 10.. These ages were selected based on our earlier findings that showed each age group was associated with maturational changes w17,18x. Experiments were carried out on Yorkshire pigs Ž Sus scrofa. ranging from 2.3 to 8.3 kg in weight. In each experiment, two litter mates were prepared simultaneously. One member of the pair was assigned randomly to the experimental group and the other to the control group. Animals were anesthetized with Saffan given i.m. Ž1 ml kgy1 . so that an external jugular vein could be catheterized to allow for continuous infusion Ž12 mg kgy1 hy1 .. Next, the femoral artery of each animal was catheterized for monitoring abdominal aorta blood pressure ŽAoP. using a Statham P23 pressure transducer and for periodic withdrawal of blood for measurements of gas tensions and pH ŽRadiometer ABL 5.. A femoral vein was then catheterized for continuous infusion of 5% dextrose in 0.9% NaCl Ž1–2 ml kgy1 hy1 . in order to maintain hydration. The trachea was cannulated, decamethonium bromide Ž100 ml kgy1 i.v.. was administered for paralysis and artificial ventilation begun with 100% O 2 . Ventilatory parameters, stroke volume and rate, were similar for both animals, thereby insuring that pH, pCO 2 and pO 2 were maintained within normal limits. Intratracheal pressure ŽITP. was monitored by a pressure transducer ŽGrass. connected to the expiratory port of the tracheal cannula. Rectal temperature was monitored and maintained at 39 " 0.58C using a servo-controlled heating blanket.

2. Materials and methods

2.2. Experimental protocol

2.1. Animals and surgical procedures

Before initiating the experimental protocol, blood gas tensions and pH were measured and brought to within normal limits, if necessary, by adjustments of ventilator rate and volume, and by infusions of NaHCO 3 . However, once the protocol was begun, no further adjustments were made. The infusion rate of Saffan anesthesia was reduced to 6 mg kgy1 hy1 Ži.v.. after the completion of surgical procedures. One hour was allowed to elapse before the initiation of the experimental protocol to allow for stabi-

This study was approved by the institutional animal care and utilization committee. All experimental procedures met with federal and state regulations and also with the ‘‘Guide for the Use and Care of Laboratory Animals’’ approved by the council of the American Physiological Society. The swine used in this study were born and nurtured by their respective sows in the Division of Labo-

Table 1 Mean arterial pressures and heart rates during baseline condition Mean arterial pressure ŽMAP. and mean heart rate ŽMHR. are presented for animals in three age groups. Data from experimental and control animals were combined within each group. Variable Žmean " S.D.. MAP ŽmmHg. MHR ŽHz. a

Age group 1 Ž5–8 days, n s 8. 76 " 20.8 3.9 " 0.72

Age group 2 significantly greater than age group 1. Age group 3 significantly less than age groups 1 and 2.

b

Age group 2 Ž13–15 days, n s 8. 99 " 18.5 4.1 " 0.70

a

Age group 3 Ž26–34 days, n s 10. 90 " 20.9 3.1 " 0.51b

108

A.L. Sica et al.r Brain Research 837 (1999) 106–116

lization. Fifteen minutes prior to the onset of hypercapnia arterial blood was withdrawn from each animal for measurement of pH and blood gas tensions. The experimental animal was then switched from breathing 100% O 2 to 10% CO 2 Žbalance O 2 ., while the control animal continued to breathe 100% O 2 . The ITP signal of the experimental animal was carefully monitored so that inflation pressures of the different gas mixtures were equivalent. In this way, the activities of slowly adapting pulmonary stretch receptors were kept constant across the two conditions. The exposure to hypercapnia lasted for 1 h, after which the

experimental animal was returned to breathing 100% O 2 for 2 h. The level of CO 2 , and duration of exposure were similar to those used to elicit FOS expression in adult animals w35x. A 2 h recovery period was established because our previous studies showed that 2 h was necessary for maximal FOS expression w33,34x. 2.3. Immunocytochemistry The immunocytochemical technique used in these experiments was identical to that described in our previous

Fig. 1. The time course of changes in pCO 2 Žtop. and pH Žbottom. are plotted for experimental animals in each of three age groups. Each data point represents the mean " S.E. Measurements were made at baseline ŽBL, y15 min. before the onset of hypercapnia Žat time 0 min., every 15 min after the onset of hypercapnia Ž15–60 min., and at the end of each recovery hour Ž120 and 180 min.. Asterisks indicate a statistically significant difference between values during hypercapnia and those of either baseline or recovery. ‡ indicates that the pH of 5–8 day old and 13–15 day old experimental animals was significantly different from baseline. q indicates that the pH of 5–8 day old animals was significantly different from baseline at the end of the recovery period.

A.L. Sica et al.r Brain Research 837 (1999) 106–116

reports w33,34x. Immediately after the conclusion of the experimental protocol, the infusion rate of Saffan was raised to 12 mg kgy1 hy1 and a wide thoracotomy performed on experimental and control piglets. Animals were then perfused transcardially with physiological saline followed by 4% paraformaldehyde in 0.1 M phosphate buffer, pH 7.4. Variability was minimized by simultaneous processing of tissues from experimental and control animals. The medullae of each pair were removed and post-fixed for 2 to 3 h in individual vials containing 4% paraformaldehyde in 0.1M phosphate buffer ŽPBS, pH 7.4.. Tissue blocks were immersed for 18–24 h at 48C in a 10% solution of sucrose dissolved in 0.1 M PBS, and frozensectioned at 35 mm. Sections were saved at 105 mm intervals, placed in spot-test wells filled with 0.1 M PBS ŽpH 7.4., washed in Tris-buffered saline ŽTBS. and pre-incubated for 30 min in goat serum, diluted 1:30 in TBS. Tissues were incubated in a commercial polyclonal anti-

109

body judged to be specific to a synthetic N terminal fragment of human FOS ŽOncogene Science. and diluted 1:1000 in a TBS solution containing 1% goat serum and 0.2% Triton-X 100. Next, tissues were incubated in a biotinylated goat anti-rabbit IgG secondary antibody Ž1:200, 30–45 min. and avidin–biotin peroxidase complex Ž1:100, 60 min. ŽVector Labs, Elite Kit.. The bound peroxidase immunoreaction product was visualized by treating tissues with a 0.05% substrate solution of the chromogen, diaminobenzidine and 0.01% H 2 O 2 in TBS. Tissues were cleared in xylene and mounted on gelatincoated slides. Data were plotted using a camera lucida attached to an Olympus BX-40 light microscope, and photographed on Kodak TMax 100 film. In order to standardize sampling, counts of neurons were taken from a total of 12 sections, each section separated by a 105 mm interval. These tissues were representative of FOS expression in brainstem regions extending through the AP and

Fig. 2. The percent difference mean arterial pressure Ž% DIF MAP. is plotted for experimental and control groups according to age, indicated above each graph. Data are plotted as the mean " S.E., with values at baseline ŽBL. set to zero. Measurements were made at BL Žy15 min. before the onset of hypercapnia Žat time 0 min., every 15 min after the onset of hypercapnia Ž15–60 min., and at the end of each recovery hour Ž120 and 180 min.. Asterisks indicate a statistically significant difference between values of control and experimental animals at designated time points.

110

A.L. Sica et al.r Brain Research 837 (1999) 106–116

caudal two-thirds of the NTS in each animal. The total number of cells that expressed FLI in the AP and NTS were counted in both experimental and control animals.

2.4. Data recording During each experiment, AoP, ECG Žbandpass filtered from 0.3 Hz to 3 kHz., and ITP signals of both animals were continuously displayed on a polygraph ŽSensormedics. to monitor physiological status. Analog signals were stored on VCR tape using an interface ŽInstrutech. for later off-line analysis. AoP, HR, pH and blood gas tensions were measured at similar intervals for both control and experimental animals during each experiment: Ža. baseline conditions, 15 min prior to onset of hypercapnia, while both animals breathed 100% O 2 ; Žb. every 15 min of the 1 h exposure to hypercapnia; Žc. at the end of each hour of a 2 h recovery period. AoP was measured during stable epochs at each of the above intervals, and was the average of four measurements, with each measure separated from the next by 15 s intervals. As described in our earlier publications, mean HR was derived from the autopower spectrum constructed from the AoP signal, thereby allowing for computation of HR over a large number of cardiac cycles w38,39x.

3. Results Successful experiments were carried out on 13 pairs of age- and litter-matched neonatal swine. There was no systematic bias in the study design due to the relative distribution of males and females within control and experimental groups Ž x 2 test, not significant.. 3.1. Age-related changes in physiological measures during baseline and hypercapnia To determine whether age-related changes were present in MAP and HR during baseline conditions, data of experimental and control animals in each age group were combined. Animals 13–15 days old exhibited a statistically significant increase of MAP compared to 5- to 8-day-old animals, and 26- to 34-day-old animals showed a statistically significant decrease in HR compared to 5- to 8-dayold and 13- to 15-day-old animals Žsee Table 1.. These results confirmed those of earlier studies w17,36x.

2.5. Data analysis To determine whether age-dependent changes were present at baseline, MAP and mean HR of control and experimental animals were combined in each age group and statistical comparisons were made using a one-way ANOVA. To compare changes of arterial blood gas tensions and pCO 2 in experimental animals a mixed-model ANOVA was used with age as an independent factor and sample time as a repeated factor. The finding of a significant main effect by ANOVA was further evaluated using post hoc Fisher pair-wise comparisons. The level for statistical significance was established at p - 0.05 for all comparisons. Data with non-normal distributions were analyzed for statistical significance using non-parametric tests. MAP and HR were expressed a percent difference from baseline and comparisons were made between control and experimental animals at each sampling time using the Mann– Whitney test. Counts of neurons expressing FOS in the AP and NTS were obtained for control and experimental animals and categorized according to treatment group and age. Comparisons of neurons expressing FOS in experimental and control animals were made using the Mann– Whitney test. Age-related comparisons were made using the Kruskal–Wallis test.

Fig. 3. Camera lucida drawings of the porcine medulla oblongata demonstrate c-fos gene induction in the nucleus of the solitary tract ŽNTS. at equivalent anteroposterior levels just rostral to calamus scriptorius in 7 ŽA., 14 ŽB. and 33 ŽC. day old stimulated piglets. Nuclear labeling occurs primarily in the dorsal region subjacent to nucleus gracilis Žng. and within medial subnuclei. The NTS of the postnatal day ŽPD. 14 neonate ŽB. demonstrates a higher density of immunolabeled nuclei than on other postnatal days ŽA, C.. Abbreviations: apc, area postrema, caudal pole; ng, nucleus gracilis; tr, solitary tract; X, vagal motor nucleus; XII, hypoglossal nucleus. Each symbol Žfilled circle. represents one immunolabeled nucleus.

A.L. Sica et al.r Brain Research 837 (1999) 106–116 Fig. 4. Photomicrographs of transverse sections of the medulla oblongata of a pair of 26 day old stimulated and control piglets. ŽA. Level of the calamus scriptorius demonstrates neurons induced to express FLI in dorsal and intermediate subnuclei of nucleus tractus solitarii ŽNTS. in a piglet exposed to CO 2 . The immunolabeled nuclei surround the solitary tract and extend dorsomedially. ŽB. Higher power view of the labeled cell group shown in ŽA., surrounding the solitary tract Žunlabeled ovoid area.. Significantly larger numbers of NTS neurons contain FLI after hypercapnia ŽB. than in the equivalent locus in the unexposed control piglet ŽC.. Arrows point to the same loci in NTS for orientation. ŽD. Rostral level of the dorsal subnucleus of NTS at level of area postrema of a 14 day old piglet. Note higher densities of cells containing FLI in the exposed animal than in ŽE. the equivalent locus in the NTS of the age-matched control. Bar s150 mm ŽA.; 50 mm ŽB–E.. 111

112

A.L. Sica et al.r Brain Research 837 (1999) 106–116

Fig. 1 depicts the time course of changes in pCO 2 Žtop graph. and pH Žbottom graph. for animals in age groups 1–3 at baseline, during exposure to 1 h of 10% CO 2 , and during 2 h of recovery. Hypercapnic stimulation produced a clear respiratory acidosis with marked decreases and increases in values of pH and pCO 2 , respectively. ŽThe pO 2 of experimental animals was ) 100 Torr during baseline, hypercapnic exposure, and recovery.. The changes in pH and pCO 2 during hypercapnia were significantly different from values at baseline. During recovery, the pH of 5–8 days old and 13–15 days old experimental animals was significantly different from baseline even 1 h after the offset of hypercapnia Žindicated by ‡ in Fig. 1 at 120 min.. The pH of 5- to 8-day-old animals, although within normal limits Ži.e., 7.35–7.45. was also significantly different from baseline at the end of the recovery period Žindicated in Fig. 1 by ‡ at 180 min.. As is apparent in Fig. 1, there were no significant age-related differences in pH and pCO 2 of experimental animals at any time point during the experiment. The values of pH and arterial blood gas tensions for control animals remained within normal experimental limits for the entire experiment: pH ranging from 7.41 to 7.43, pCO 2 from 34 to 38 Torr, and pO 2 ) 100 Torr. Comparisons of percent difference MAP for control and experimental animals did not reveal a significant overall treatment effect for prolonged exposure to hypercapnia. However, examination of Fig. 2 Žtop graph. shows that a significant increase in MAP was present in 5- to 8-day-old animals after 15 min of hypercapnia, and also at the end of the recovery period; intervening time points did not differ significantly from values of control animals. The early increase of MAP, shown by 5- to 8-day-old experimental animals, was also significantly different from that of 13- to 15-day-old experimental animals Žnot shown.. The middle graph of Fig. 2 shows no change in MAP as a result of hypercapnia in the 13- to 15-day-old group compared to same aged animals in the control group. Surprisingly, the spontaneous changes in MAP of the 26- to 34-day-old control group ŽFig. 2, bottom graph. were actually greater than those of the experimental group, with significance attained for all time points beginning at 45 min after the onset of hypercapnia. This pattern of change in MAP also distinguished 26- to 34-day-old control animals from 13to 15-day-old control animals. The former group had increases of MAP after 60 min of hypercapnia and at the end of the first hour of recovery that were significantly greater than the MAP of the latter group Žnot shown.. There were no statistically significant differences in percent difference HR between control and experimental animals. Neither was there any difference between HR values of experimental animals in different age groups. 3.2. Hypercapnic-induced FOS expression in NTS and AP FOS-like immunoreactive nuclei were present in the NTS and AP of both experimental and control animals.

The presence of FOS- like immunoreactivity ŽFLI. in the latter group gave a measure of constitutive expression, i.e., the amount of FLI expressed as a result of procedures, e.g., anesthesia, surgery, that were experienced by both groups of animals. Any subsequent change in FLI of experimental animals can be relegated to the experimental manipulation, in this case, exposure to 10% CO 2 . NTS neurons expressing FLI after 1 h of exposure to 10% CO 2 were distributed topographically and far outnumbered those observed in age-matched controls Žsee Figs. 7 and 8.. FLI was restricted to neurons localized to

Fig. 5. Camera lucida drawings of transverse sections of porcine medulla oblongata rostral to obex compare FOS-like immunoreactivity ŽFLI. in the area postrema Žap. and subjacent sub-postrema zone at postnatal day ŽPD. 26. ŽA. Markedly larger number of neurons contain FLI in the ap and subjacent zone of this CO 2 -stimulated piglet as compared to ŽB. the control. Each symbol Žfilled circle. represents one immunolabeled nucleus.

A.L. Sica et al.r Brain Research 837 (1999) 106–116

113

Unlike the NTS, AP neurons did not exhibit a topographic pattern of expression. The number of AP neurons with FLI in experimental animals was markedly greater than the degree of constitutive expression in control animals. Figs. 5 and 6 demonstrate c-fos gene expression in the AP in response to hypercapnia at postnatal day 26. Statistical comparisons were made of the total number of neurons expressing FOS in the NTS and AP of control and experimental animals. In Fig. 8A, neurons with FLI in each age group were combined, and demonstrated that both the NTS and AP of experimental animals exhibited statistically significant increased FOS expression compared to control animals. Age-related differences in numbers of neurons with FLI were noted in the NTS of both experimental and control animals; no such differences, however, were observed in the AP Žsee Figs. 7 and 8B.. In the experimental group, significantly greater numbers of NTS neurons with FLI were exhibited by age group 2 and 3 animals compared to age group 1 animals ŽFig. 7, left panel.. However, it should be noted that those neuronal counts were not adjusted for constitutive expression, which also differed in degree of FOS expression. For example, more neurons with FLI were noted in the NTS of age group 2 and 3 animals than in the NTS of age group 1 animals ŽFig. 7, left panel: control group.. To determine whether increases in FOS expression were related to age of experimental animals, the total number of neurons expressing FLI in each animal was adjusted by subtracting the number of neurons expressing FLI in the paired-control Fig. 6. Photomicrographs of porcine area postrema Žap. on transverse sections of the medulla at levels rostral to obex in a 14 day old pair of stimulated ŽA. and control ŽB. piglets. Note larger numbers of ap neurons induced to express FLI in the piglet exposed to CO 2 as compared to the age-matched control. Subjacent to the ap is an elongate strip of nuclei containing FLI within the sub-postrema subnucleus of NTS Žsee text for details.. Bar s90 mm.

specific subnuclei of the caudal two thirds of NTS, corresponding to the general visceral afferent division. Large numbers were present within the medial and sub-postrema divisions, where labeled cells were concentrated medial to the solitary tract. A prominent sheet of immunolabeled nuclei corresponding to the dorsal subnucleus of the NTS was located at the level of the calamus scriptorius, and was subjacent to the nucleus gracilis ŽNG.. No immunoreactive neurons were detected in the NG, or adjoining nuclei, such the nucleus hypoglossi. A distinct population of labeled cells abutted the lateral border of the dorsal motor nucleus. Fig. 3 shows the distribution and density patterns in the NTS and dorsal vagal motor nucleus of three representative neonates at different stages of development, with the largest number of neurons with FLI in the NTS of the 14-day-old piglet. Induction patterns in representative pairs of age-matched piglets are shown in Fig. 4.

Fig. 7. Total number of neurons Ž"S.E.. with FOS-like immunoreactivity ŽFLI. in the nucleus of the solitary tract ŽNTS; left panel. and area postrema ŽAP; right panel. of experimental and control animals. Left panel — Asterisks indicate that NTS neurons of control animals in age groups 2 and 3 expressed significantly more constitutive FLI than age group 1 animals. Experimental animals in age groups 2 and 3 exhibited more neurons in the NTS with FLI than age group 1 animals Žindicated by stars.; the q symbol indicates more neurons with FLI in the NTS of age group 2 animals than age group 3 animals. Right panel — AP neurons show no age-related differences in expression of FLI.

114

A.L. Sica et al.r Brain Research 837 (1999) 106–116

Fig. 8. ŽA. Total number of neurons Ž"S.E.. with FOS-like immunoreactivity ŽFLI. in the nucleus of the solitary tract ŽNTS; left panel. and area postrema ŽAP; right panel.. Data from each age group have been combined. Asterisks indicate that experimental animals have significantly greater numbers of neurons with FLI than control animals. ŽB. Adjusted number of counts Ž"S.E.., i.e., total count of AP and NTS neurons with FLI in each experimental animal minus total count of paired-control animal. Left panel, adjusted number of neurons expressing FOS in the NTS of experimental animals in each of three age groups. The asterisk indicates a significant increase of FLI in age group 2. Right panel, adjusted number of counts for AP neurons expressing FOS in each of three age groups.

animal, thereby removing the effect of constitutive FOS expression. This procedure revealed an age-related increase of FOS expression in the NTS of age group 2 animals ŽFig. 8B, left panel.; the AP, in contrast, revealed no age-relationships ŽFig. 8B right panel..

4. Discussion The results of the present investigation are the first to describe hypercapnic-induced changes in FOS expression of the AP and NTS in a developing animal model. The comparison of NTS and AP neurons revealed that hypercapnia, a relatively simple stimulus, evoked patterns of FOS expression that differed with respect to topographic organization and to developmental stage. NTS neurons exhibited both topographical organization and a relation-

ship between maturation and number of neurons expressing FOS. In contrast, AP neurons failed to demonstrate either topography or a relationship with maturation. The overall topographic organization of NTS neurons in the piglet was similar to that of mature animals of other species w19,22,26,40x. While hypercapnic-induced expression of FOS is not selective for neurons subserving any particular function, the overall pattern of such expression suggests some neurons with FLI are involved in processing peripheral chemoreceptor and baroreceptor information, and possibly in shaping central respiration- and sympathetic-related activities. For example, that neurons with FLI in the dorsal strip and medial subnucleus of the NTS in the piglet might participate in processing peripheral chemo- and baroreceptor information is supported by studies in the rat and cat showing that these afferent inputs terminated in those regions w6,11,14,24x. Furthermore, peripheral chemoreceptors have been shown to increase impulse activity during hyperoxic–hypercapnic stimulation, but not to the high discharge rates observed during hypoxic– or asphyxic–hypercapnia w15x. Thus, because the relationship between alterations of impulse activity along afferent pathways and FOS expression in recipient neurons is unknown, a contribution of peripheral chemoreceptors to the pattern of FOS expression could not be discounted. However, there is evidence that the pattern of FOS expression in the NTS of piglets might represent more the contribution of central CO 2-chemoreceptors than peripheral chemoreceptors because similar patterns were noted in the NTS of peripherally-chemodenervated adult animals of other species w19,21x. While the AP exhibited no age-related changes in number of neurons with FLI, neurons of the NTS showed age-related increases in FOS expression. However, this increase followed a biphasic pattern: NTS neurons with FLI increased in number over the first two postnatal weeks but decreased in number during the fourth and fifth weeks, to levels noted during the first week. Parenthetically, this observation suggested that two week old animals represented a developmental stage characterized by central hyper-excitability to afferent stimulation. This interpretation is also supported by our finding that even constitutive FOS expression was greatest in two week old animals. Thus, this time period may represent a vulnerable stage of development in swine since chemodenervation either alone or in combination with barodenervation was associated with potentially fatal apneas in the former case, and with prolonged apneas and death in the latter w10,12x. In this study, we attempted to establish a relationship between level of FOS expression and change in MAP and HR; no such relationship was found. Such absences of any associations between anatomical and physiological variables may be partially due to the differential effects of anesthesia on developing animals. However, we have shown that Saffan at the infusion rates used in this study maintained an anesthetic state characterized by EEG spec-

A.L. Sica et al.r Brain Research 837 (1999) 106–116

tral components distributed across the delta Ž0.5–4.0 Hz. and theta Ž4–7 Hz. bands regardless of differences in animals ages and weights w37x. Also, Saffan is known to preserve cardiorespiratory reflex activity in developing swine w18,20x. Hence, we do not believe that our findings were confounded by age-related differences in responses to anesthesia. We may speculate that the absence of the expected increase in MAP and HR is itself a response to prolonged hypercapnia. This expectation of age-related increases of MAP and HR was based on the results of earlier studies that reported such increases in piglets up to 4 weeks of age w17,32x. However, age-related comparisons were not made in those studies, and the duration of C0 2 exposure was shorter than that used in the present investigation. Such different results highlight the fact that hypercapnic stimulation may elicit non-stereotypical responses with patterns that represent maturational status and stimulus parameters, i.e., exposure duration and concentration of CO 2 . The possibility that changes in MAP were modulated by hypercapnic-induced changes in brainstem respiratory neurons was not evaluated in this study. Such an assessment would have been valuable, but it should be noted that the response of brainstem respiratory neurons to prolonged hypercapnia has not been described in the developing animal. Cardiovascular responses to long-term exposure to hypercapnia may elicit central mechanisms in the neonate that tend to maintain MAP and HR at relatively constant levels. This may constitute an important defense as neonates may be unable to mount sustained increases of cardiovascular activity during prolonged chemoreceptor activation due to insufficient metabolic andror neurotransmitter stores. One mechanism that may be involved in maintaining MAP at a relatively constant level has been suggested by studies of mature animals of other species. These studies demonstrated a contribution by the AP to baroreceptor-related sympathoinhibition via connections with the NTS w4,5x. If such projections develop function during the postnatal period, then, it is possible that APmediated sympathoinhibition was low during the first week of life, allowing for some increase of MAP, and that such inhibition became more pronounced in older animals so that changes of MAP were minimal. This idea is supported by our observation of the trend for the total number of neurons with FLI to be lowest in the AP and NTS of 5- to 8-day-old animals Žsee Fig. 8., the group that demonstrated an early increase of MAP during hypercapnia. In addition, the finding that MAP was either unchanged or exhibited less variation than age-matched controls Ž13- to 15-day-old and 26- to 34-day-old piglets, respectively. could also provide support for a mechanismŽs. that maintained MAP relatively constant during prolonged, severe hypercapnia. Hence, it is plausible that the absence of a sustained change in MAP andror HR is a response type, possibly characteristic of neonates. Such a maneuver would also conserve cellular metabolic andror neurotransmitter stores,

115

and thereby prevent loss of normal regulation of cardiovascular-related activity. Acknowledgements We wish to thank I. Frasier, M. Anwar, C. Wong and S. Ingenito for their excellent technical assistance. This work was supported by National Institutes of Health grants: HL-20864 ŽPMG, ALS, DAR., HD-28931ŽPMG, ALS., NS-36363 ŽDAR.. References w1x D.G. Bernard, A. Li, E.E. Nattie, Evidence for central chemoreception in the midline raphe, J. Appl. Physiol. 80 Ž1996. 108–115. w2x F. Bongianni, D. Mutolo, M. Carfi, T. Pantaleo, Area postrema receptors mediate respiratory and gastric responses in the rabbit, Neuroreport 9 Ž1998. 2057–2062. w3x H.L. Borison, R. Borison, L.E. McCarthy, Role of the area postrema in vomiting and related functions, Fed. Proc. 43 Ž1984. 2955–2958. w4x Y. Cai, M. Hay, V.S. Bishop, Synaptic connections and interactions between area postrema and nucleus tractus solitarius, Brain Res. 724 Ž1996. 121–124. w5x C.-Y Chen, A.C. Bonham, Non-NMDA and NMDA receptors transmit area postrema input to aortic baroreceptor neurons in NTS, Am. J. Physiol. 275 Ž1998. H1695–H1706. w6x J. Ciriello, F.R. Calaresu, Projections from buffer nerves to the nucleus of the solitary tract: an anatomical and electrophysiological study in the cat, J. Auton. Nerv. Syst. 3 Ž1981. 299–310. w7x B.A. Chiodini, B.T. Thach, Impaired ventilation in infants sleeping facedown: potential significance for sudden infant death syndrome, J. Pediatr. 123 Ž1993. 686–692. w8x E.L. Coates, A. Li, E.E. Nattie, Widespread sites of brainstem ventilatory chemoreceptors, J. Appl. Physiol. 75 Ž1993. 5–14. w9x H.M. Coleridge, J.C.G. Coleridge, D. Jordan, Integration of ventilatory and cardiovascular control systems, in: R.G. Crystal, J.B. West et al. ŽEds.., The Lung: Scientific Foundations, 2nd edition, Lippincott-Raven, Philadelphia, 1997, pp. 1839–1849. w10x A. Cote, H. Porras, B. Meehan, Age-dependent vulnerability to carotid chemodenervation in piglets, J. Appl. Physiol. 80 Ž1996. 323–331. w11x S. Donague, R.B. Felder, D. Jordan, K.M. Spyer, The central projections of carotid baroreceptors and chemoreceptors in the cat: a neurophysiological study, J. Physiol., Lond. 347 Ž1984. 397–409. w12x D.F. Donnelly, G.G. Haddad, Prolonged apnea and impaired survival in piglets after sinus and aortic nerve section, J. Appl. Physiol. 68 Ž1990. 1048–1052. w13x R.D. Fields, Signaling from neural impulses to genes, The Neuroscientist 2 Ž1996. 315–325. w14x J.C. Finley, D.M. Katz, The central organization of carotid body afferent projections to the brainstem of the rat, Brain Res. 572 Ž1992. 108–116. w15x R.S. Fitzgerald, D.C. Parks, Effect of hypoxia on carotid chemoreceptor response to carbon dioxide in cats, Respir. Physiol. 12 Ž1971. 218–229. w16x A. Frugiere, E. Nunez, R. Pasaro, S. Gaytan, J.C. Barillot, Efferent projections from the rostral ventrolateral medulla to the area postrema in rats, J. Auton. Nerv. Syst. 72 Ž1998. 34–45. w17x P.M. Gootman, N.M. Buckley, N. Gootman, Postnatal maturation of neural control of the circulation, in: E.M. Scarpelli, E.V. Cosmi ŽEds.., Reviews in Perinatal Medicine, Raven Press, New York, 1979, pp. 1–71. w18x P.M. Gootman, H.L. Cohen, A.M. Steele, A.L. Sica, G. Condemi,

116

w19x

w20x

w21x

w22x

w23x

w24x

w25x

w26x

w27x

w28x

w29x

A.L. Sica et al.r Brain Research 837 (1999) 106–116 M.R. Gandhi, L.T. Eberle, Effect of anesthesia on efferent phrenic activity in neonatal swine, Brain Res. 522 Ž1990. 131–134. M.A. Haxhiu, K. Yung, B. Erokwu, N.S. Cherniack, CO 2-induced c-fos expression in the CNS catecholaminergic neurons, Respir. Physiol. 105 Ž1996. 35–45. S.M. Hilton, J.M. Marshall, The pattern of cardiovascular responses to carotid chemoreceptor stimulation in the cat, J. Physiol., Lond. 326 Ž1982. 495–513. A.H. Jansen, D.M. Nance, P. Liu, H. Weisman, V. Chernick, Effect of sinus denervation and vagotomy on c-fos expression in the nucleus tractus solitarius after exposure to CO 2 , Pflugers Archives — The European Journal of Physiology 431 Ž1996. 876–881. N. Larnicol, F. Wallois, P. Berquin, F. Gros, D. Rose, C-fos-like immunoreactivity in the cat’s neuraxis following moderate hypoxia or hypercapnia, J. Physiol., Paris 88 Ž1994. 81–88. A. Li, E.E. Nattie, Focal central chemosensitivity in the RTN studied with a CO 2 diffusion pipette in vivo, J. Appl. Physiol. 83 Ž1997. 420–428. J. Lipski, R.M. McAllen, A. Trzebski, Carotid baroreceptor and chemoreceptor inputs onto single medullary neurones, Brain Res. 107 Ž1976. 132–136. M. Maskrey, Influence of body temperature on responses to hypoxia and hypercapnia: implications for SIDS, Clin. Exp. Pharmacol. Physiol. 22 Ž1995. 527–532. M. Miura, J. Okada, K. Takayama, T. Suzuki, Neuronal expression of FOS and JUN protein in the rat medulla and spinal cord after anoxic and hypercapnic stimulations, Neurosci. Lett. 178 Ž1994. 227–230. S. Mosko, C. Richard, J. McKenna, S. Drummond, D. Mukai, Maternal proximity and infant CO 2 environment during bedsharing and possible implications for SIDS research, Am. J. Phys. Anthropol. 103 Ž1997. 315–328. E.E. Nattie, A. Li, Central chemoreception in the region of the ventral respiratory group in the rat, J. Appl. Physiol. 81 Ž1996. 1987–1995. J. Pineda, G.K. Aghajanian, Carbon dioxide regulates the tonic activity of locus ceruleus neurons by modulating a proton- and

w30x

w31x

w32x

w33x

w34x

w35x

w36x

w37x

w38x

w39x w40x

polyamine-sensitive inward rectifier potassium current, Neuroscience 77 Ž1997. 723–734. A.L. Ponsonby, T. Dwyer, L.E. Gibbons, J.A. Cochrane, Y.G. Wang, Factors potentiating the risk of sudden infant death syndrome associated with the prone position, N. Engl. J. Med. 329 Ž1994. 377–382. Z.M. Qian, H.W. Koon, Area postrema is essential for the maintenance of normal blood pressure under cold stress in rats, Exp. Brain Res. 121 Ž1998. 186–190. G.D. Reddy, N. Gootman, P.M. Gootman, L. Crane, Regional blood flow changes in neonatal pigs in response to hypercapnia, hemorrhage and sciatic nerve stimulation, Biol. Neonate 25 Ž1974. 249– 262. D.A. Ruggiero, S. Tong, N. Gootman, P.M. Gootman, Hypotension-induced expression of the c-fos gene in the medulla oblongata of piglets, Brain Res. 706 Ž1996. 199–209. D.A. Ruggiero, A.L. Sica, M. Anwar, I. Frasier, N. Gootman, P.M. Gootman, Induction of c-fos gene expression by spinal cord transection in Sus Scrofa, Brain Res. 759 Ž1997. 301–305. J.W. Sato, J.W. Severinghaus, A.I. Basbaum, Medullary chemoreceptor CO 2 neuron identification by c-fos immunocytochemistry, J. Appl. Physiol. 73 Ž1992. 96–100. B.S. Segal, J.D. Inman, I.R. Moss, Respiratory responses of piglets to hypercapnia during postnatal development: effect of opioids, Ped. Pulm. 11 Ž1991. 113–119. A.L. Sica, A.M. Steele, M.R. Gandhi, N. Prasad, Factors affecting central inspiratory modulation of hypoglossal motoneuron activity in newborn pigs, J. Dev. Physiol. 12 Ž1988. 285–295. A.L. Sica, Z.A. Siddiqi, Reduction of baroreceptor-related sympathetic discharge during NMDA receptor antagonism, Brain Res. 665 Ž1994. 323–326. A.L. Sica, B.W. Hundley, P.M. Gootman, Postganglionic sympathetic discharge in neonatal swine, Pediatr. Res. 39 Ž1996. 85–89. L.J. Teppema, J.G. Veening, A. Kranenburg, A. Dahan, A. Berkenbosch, C.N. Olievier, Expression of c-fos in the rat brainstem after exposure to hypoxia and to normoxic and hyperoxic hypercapnia, J. Comp. Neurol. 388 Ž1997. 169–190.