Subtle alterations in breathing and heart rate control in the 5-HT_1A receptor knockout mouse in early postnatal development

Abstract

We hypothesized that absence of the 5-HT_1A receptor would negatively affect the development of cardiorespiratory control. In conscious wild type (WT) and 5-HT_1A receptor knockout (KO) mice, we measured resting ventilation (V̇e), oxygen consumption (V̇o₂), heart rate (HR), breathing and HR variability, and the hypercapnic ventilatory response (HCVR) at postnatal day 5 (P5), day 15 (P15), and day 25 (P25). In KO mice compared with WT, we found a 17% decrease in body weight at only P5 (P < 0.01) and no effect on V̇o₂. V̇e was significantly (P < 0.001) lower at P5 and P25, but there was no effect on the HCVR. Breathing variability (interbreath interval), measured by standard deviation, the root mean square of the standard deviation (RMSSD), and the product of the major (L) and minor axes (T) of the Poincaré first return plot, was 57% to 187% higher only at P5 (P < 0.001). HR was 6–10% slower at P5 (P < 0.001) but 7–9% faster at P25 (P < 0.001). This correlated with changes in the spectral analysis of HR variability; the low frequency to high frequency ratio was 47% lower at P5 but 68% greater at P25. The RMSSD and (L × T) of HR variability were ∼2-fold greater at P5 only (P < 0.001; P < 0.05). We conclude that 5-HT_1A KO mice have a critical period of potential vulnerability at P5 when pups hypoventilate and have a slower respiratory frequency and HR with enhanced variability of both, suggesting abnormal maturation of cardiorespiratory control.

to investigate the role of the 5-HT_1A receptor in cardiorespiratory control in early postnatal development, we used the 5-HT_1A receptor knockout (KO) mouse. In these mice, the 5-HT_1A receptor is absent at both pre- and postsynaptic locations throughout the brain, and these mice display increased anxiety-like behavior in several behavioral testing paradigms in adulthood (22, 48, 55). Given the role of serotonin (5-HT) as a trophic factor in brain development, surprisingly, these mice do not display any obvious anatomical or morphological abnormalities in the brain; moreover, they grow and reproduce normally, and have normal viability (24, 48, 55). Still, 5-HT via 5-HT_1A receptors has been shown to influence cardiorespiratory control in multiple experimental paradigms. For example, overexpression of the 5-HT_1A receptor in mice results in autonomic dysfunction and mortality in early life (4). Also, 5-HT neurons in the medullary raphe are known to contribute to central chemoreception (16, 17, 25, 38), and 5-HT_1A receptor immunoreactivity has been documented in brain stem nuclei potentially involved in the central chemoresponse, suggesting a potential role for this receptor in central chemoreception (6, 42). In fact, pharmacological inhibition of 5-HT neurons by focal microdialysis of the 5-HT_1A receptor agonist, 8-hydroxy-2-(di-n-propylamino) tetralin (8-OH-DPAT), into the medullary raphe of piglets, attenuates the ventilatory response to hypercapnia at postnatal ages of 8 days or older (43). In adult rats, focal microdialysis of 8-OH-DPAT into the medullary raphe reduces the ventilatory response to hypercapnia (58), and intracisternal injection attenuates the heart rate (HR) response associated with activation of the dorsomedial hypothalamus (28). In addition, abnormalities of 5-HT_1A receptors have been reported in the sudden infant death syndrome (SIDS) (19, 40, 47, 50). Defined as the sudden and unexpected death of an infant less than 12 mo of age that is unexplained by an autopsy and death scene investigation (19, 35, 36, 60), SIDS is postulated to be due to a failure of protective cardiorespiratory responses to homeostatic stressors during a critical period (35, 36, 60). While several neurotransmitter defects have been reported in SIDS cases (9, 19, 35, 36, 50), decreased 5-HT_1A receptor binding is the most consistently reported 5-HT abnormality in the medulla (19, 40, 47, 50), with receptor deficits present both in nuclei containing 5-HT cell bodies and in nuclei receiving 5-HT projections (19, 50). In one study, the 5-HT_1A deficit was associated with an increase in immature 5-HT neurons and a relative decrease in 5-HT transporter binding (50), and, in another study, the deficit was associated with a decrease in tissue levels of 5-HT and the major enzyme involved in central 5-HT synthesis, tryptophan hydroxylase 2 (TPH2) (19). Furthermore, in a case study of an infant who died of SIDS at two postnatal weeks, subtle abnormalities in respiratory and autonomic function in this infant at two postnatal days were associated with decreased binding of tritiated lysergic acid diethylamide ([³H]LSD) to 5-HT_1A-D and 5-HT₂ receptors in the medulla (34). These findings suggest that the deficit in 5-HT_1A receptor binding is a major component of the 5-HT abnormality in SIDS, and thus an assessment of the role of the 5-HT_1A receptor in cardiorespiratory control during early postnatal development might be useful for a better understanding of the role of 5-HT_1A receptors in the pathogenesis of SIDS.

In the following study, we tested the hypothesis that the 5-HT_1A receptor is important for development of cardiorespiratory control and that absence of this receptor results in alterations in: 1) resting ventilation, 2) the ventilatory response to hypercapnia, and 3) the normal development of autonomic control in early postnatal life. To test this hypothesis, we measured breathing, oxygen consumption (V̇o₂), HR, the frequency of spontaneous bradycardias, and the variability in breathing and HR in conscious, unrestrained male and female 5-HT_1A receptor KO and wild-type (WT) mice during normoxia and during hypercapnia at postnatal day 5 (P5), day 15 (P15), and day 25 (P25).

METHODS

Ethical approval.

All experimental protocols were approved by the Institutional Animal Care and Use Committee at Dartmouth College and were in accordance with the guidelines of the National Institutes of Health for animal care.

Experimental animals.

Experiments were performed on 5-HT_1A KO and WT pups at three postnatal (P) ages (P5–6, designated P5; P14–16, designated P15; and P22–28, designated P25). Pups were obtained from 15 different litters, born to heterozygous (5-HT_1A +/−) mating pairs of the 129/SvEvTac strain (54), and consisted of 22 KO (13 males, 9 females) and 31 WT (15 males, 16 females) pups. The original breeders were obtained from the laboratory of Dr. René Hen (Columbia University, New York, NY). Forty-one pups were studied at P5 (19 KO, 22 WT), 41 pups were studied at P15 [19 KO (16 of which were also studied at P5), 22 WT (14 of which were also studied at P5)], and 41 pups were studied at P25 (19 KO, 22 WT; all of which were also studied at P15). The mice were housed in a temperature-controlled environment on a 12:12-h light-dark cycle in the Animal Resource Facility and had ad libitum access to standard rodent chow and water.

DNA preparation from tail biopsy.

A small segment of tail tissue (∼0.25 cm) was collected from all pups at P4 and incubated with 200 μl of 50 mM NaOH at 94°C for 20 min in a microcentrifuge tube. At the end of the incubation period, 50 μl of 1 M Tris·HCl, pH 8.0, were added and the contents vortexed briefly, then centrifuged at 10,000 g for 10 min. The supernatant, which contained the DNA, was collected and used to perform the PCR genotyping.

PCR genotyping.

PCR analysis was performed using Lucigen EconoTaq Plus Green 2× Master Mix (2× Taq Mix; Lucigen, Middleton, WI) and the following primer sequences (Invitrogen, Carlsbad, CA): 5HT1A promoter (XZAU): 5′-CAG TCT CTA GAT CCC CTC CCT T-3′; 5HT1A coding sequence (HTR1A): 3′-GGG CGT CCT CTT GTT CAC GTA G-5′; tTA insert (XZTL): 3′-AAG GGC AAA AGT GAG TAT GGT G-5′. A 25-μl PCR genotyping reaction was performed in a PTC-100 thermocycler (Bio-Rad Laboratories, Hercules, CA). Each PCR reaction consisted of 12.5 μl of 2× Taq Mix, 11.5 μl of primer mix (consisting of 28 μl of 1 μg/μl XZAU primer, 14 μl of 1 μg/μl HTR1A primer, 14 μl of 1 μg/μl XZTL primer, 1,144 μl of deuterated H₂O), and 1 μl of genomic DNA. The PCR cycles consisted of an initial denaturation at 95°C for 1 min, followed by 2 cycles of 95°C for 15 s, 64°C for 15 s, 72°C for 90 s; 2 cycles of 95°C for 15 s, 61°C for 15 s, 72°C for 90 s; 20 cycles of 95°C for 15 s, 58°C for 15 s, 72°C for 90 s; 10 cycles of 95°C for 15 s, 55°C for 15 s, 72°C for 90 s, and a final extension at 72°C for 10 min. A negative control amplification that did not have any DNA template was also performed. The amplified products were analyzed by gel electrophoresis at 100 mV for 45 min, using 1–2% agarose gel containing 0.5 μg/ml ethidium bromide in 1× Tris-acetate EDTA (TAE) buffer. The ethidium bromide-stained DNA was visualized using a UV transilluminator (T1201; Sigma-Aldrich, St. Louis, MO), and the gel images were captured with Alpha DigiDoc software (Alpha Innotech, San Leandro, CA). The WT and KO alleles separated on the agarose gel as 600 bp and 300 bp PCR products, respectively.

Experimental design.

Breathing was measured with either a stop-flow (P5) or flow-through (P15 and P25) whole body plethysmograph made of Plexiglas. Three different sized plethysmographs were used to accommodate the changes in body size as the pups matured. The volumes of the plethysmographs were ∼42 ml (diameter = 3 cm, length = 6 cm; P5), ∼63 ml (diameter = 4.8 cm, length = 3.5 cm; P15), and ∼118 ml (diameter = 5 cm, length = 6 cm; P25). Each plethysmograph was connected to a reference chamber of a similar size to prevent minor pressure changes in the surrounding environment from influencing the pressure in the system. Ambient temperature (TA) inside the plethysmograph was maintained at 30° C with an external heating lamp and was continuously monitored with a thermistor probe (Omega Engineering, Stamford, CT). Calibrations were made using a standard procedure of injecting 0.02 ml, 0.05 ml, or 0.1 ml of air into the smallest to largest plethysmograph, respectively, and the pressure changes associated with each injection were measured with a differential pressure transducer (Validyne Engineering, Northridge, CA). Gas bubbled in water was flushed through the plethysmograph at a flow rate of ∼60 ml/min (P5), ∼100 ml/min (P15), or ∼185 ml/min (P25), and as it exited the plethysmograph, the gas was passed through a column of drierite (Hammond Drierite, Xenia, OH) before it was sampled by an oxygen (O₂) and a carbon dioxide (CO₂) analyzer (S-3A/I, AEI Technologies, Pittsburgh, PA; and Capstar-100, CWE, Ardmore, PA, respectively). These flow rates prevented accumulation of CO₂ in the plethysmograph. HR was measured with a telemetric device (CTA-F40; DSI, St. Paul, MN) consisting of two ECG leads that were placed on the surface of the pup's chest and held in place with Ten20 conductive ECG paste (Biopac Systems, Aurora, CO). A vest was made from a cohesive flexible bandage (Andover, Salisbury, MA).

Experimental procedure.

Breathing and HR were measured during the light period in all animals at P5, P15, and P25 in room air (RA) and during hypercapnia (5% CO₂). Each pup was separated from the litter, weighed, fitted with the ECG telemetric device, and placed in the whole body plethysmograph where the pup was allowed to acclimatize in RA for 10 min or until the pup was resting quietly and breathing appeared regular.

For the P5 pups (Fig. 1, top), HR was recorded for 30 min while the plethysmograph was continuously flushed with RA, after which five 60-s periods of breathing and HR were recorded with the flow through the plethysmograph stopped. The flow through the plethysmograph was stopped so as to amplify the pressure signal associated with inspiration and expiration, because the pressure signal in the flow-through system for these tiny animals was too small to be detected. After each 60-s period of breathing and HR recording, the plethysmograph was flushed with RA for 60 s. Gas flow through the plethysmograph was then switched to 5% CO₂-21% O₂. HR was recorded for 5 min during continuous hypercapnia, followed by five 60-s recordings of breathing and HR when flow through the plethysmograph was stopped. As in RA, the plethysmograph was flushed with 5% CO₂-21% O₂ for 60 s between each 60-s breathing and HR recording.

Fig. 1.

Schematic of the experimental protocol employed to assess breathing and heart rate (HR) responses to hypercapnia in 5-HT_1A knockout (KO) and wild type (WT) pups at postnatal day 5 (P5), postnatal day 15 (P15), and postnatal day 25 (P25). In P5 pups (top), the flow through the plethysmograph was stopped for 1 min at a time after the 30 min and the 5 min HR recordings in room air (RA) and 5% CO₂, respectively, to amplify the signal associated with inspiration and expiration, whereas breathing and HR were recorded continuously in the older pups (bottom).

For the P15 and P25 pups (Fig. 1, bottom), breathing and HR in RA were recorded continuously for 30 min, after which the gas was switched to 5% CO₂-21% O₂, and breathing and HR were again recorded continuously for 15 min. The O₂ and CO₂ concentrations of the gas exiting the plethysmograph during RA and hypercapnia were also recorded. At the end of the experiment, the rectal temperature of each pup was measured with a fine thermocouple (Omega Engineering, Stamford, CT), and then the pup was returned to the litter. The barometric pressure on each experimental day was also recorded with a workstation digital barometer (Control, Friendswood, TX).

Data analysis.

We were unable to test the same mice at all three experimental ages, because 12 pups (3 of 22 KOs and 9 of 31 WTs) died spontaneously in their home cages between P6 and P15 during the course of this study. The cause of this unexpected mortality is unknown. LabChart6 (AD Instruments, Colorado Springs, CO), a digital strip chart recorder that also facilitates data calculations, was used to obtain and analyze the data. The last 5 min of the breathing tracings during both gas conditions was used to calculate minute ventilation (V̇e; ml·min⁻¹·g⁻¹). Breathing frequency (f; breaths/min) and the amplitude of the pressure deflections associated with inspiration and expiration were obtained directly from the breathing traces. The amplitude of the pressure deflections associated with inspiration and expiration, along with the plethysmograph and rectal temperatures, calibration volume, amplitude of the pressure deflection associated with the calibration volume and the barometric pressure, were used to calculate the tidal volume (V_T; ml/g) from the equation derived by Drorbaugh and Fenn (15, 18). The product of f and V_T was used to estimate V̇e. V̇o₂ in normoxia and during hypercapnia was calculated as the product of the gas flow rate and the difference between the inspired and mixed expired O₂ and then normalized to body weight {[flow (ml/min)·(FIO₂ − FEO₂)]/body wt (g)}. The ratio of V̇e/V̇o₂ was also calculated for each pup during both gas conditions.

HR (beats per minute; bpm) was obtained directly from the last 5 min of the ECG recordings during each gas condition. The entire ECG recording in normoxia (30 min) and hypercapnia (15 min) was used to assess the frequency of spontaneous bradycardias. A bradycardia was arbitrarily defined as a 5% or greater decrease in HR from baseline lasting for more than two cardiac cycles. The frequency (min⁻¹) and severity [duration (milliseconds) and magnitude of the HR change (bpm)] of bradycardias were determined in normoxia and hypercapnia.

Custom-designed analysis software was used to assess breathing and HR variability (HRV) in the time and frequency domains. Interbreath intervals (IBI; s⁻¹) and the R-wave-to-R-wave intervals (RRI; s⁻¹) from three 1-min segments of continuous, nonoverlapping breathing and ECG tracings sampled at 1 kHz were collected during normoxia and during hypercapnia. The root of the mean square of successive differences (RMSSD) and the standard deviation (SD) as well as the elliptical approximation to the Poincaré first return plot (major axis, L, minor axis, T), and an estimate of the area of the Poincaré distribution (L × T) were computed and used to evaluate the nature of interbeat/breath variability (10). HRV in the frequency domain was estimated using the Lomb spectral analysis technique as proposed by Moody (44), to avoid the spectral distortion caused by resampling of the original RRI time series. Spectral power between limits proposed by Just et al. (32) were applied to the RRI power spectral density to estimate low frequency (LF) band power (0.15–1.5 Hz) and high frequency (HF) band power (1.5–5.0 Hz). The RRI LF/HF ratio was computed as the ratio of the respective power band values. The HF band is thought to reflect parasympathetic inputs to HR control, whereas the LF band is thought to reflect a combination of sympathetic and parasympathetic inputs to HR control; the LF/HF ratio is meant to capture the balance of parasympathetic and sympathetic activity (1, 41).

Statistical analysis.

The data are presented as means ± SD. Kaplan Meier survival analysis was used to compare the cumulative survival at the endpoint of the study at P25, only between those KO and WT pups that were tested from the start of the study at P5. Pups that were added at time points later than P5 were excluded from this survival analysis. The effects of gender and genotype on body weight were assessed by two-way analysis of variance (ANOVA) at each age. The effects of gender and genotype on breathing, HR, the occurrence of spontaneous bradycardias, and variability in breathing and HR during normoxia and hypercapnia were assessed by two-way repeated measures ANOVA at each age with gas (normoxia or hypercapnia) as the repeated factor. When there was no gender effect, the data for both genders were combined. Differences were considered significant when P < 0.05. Bonferroni post hoc comparisons were made when significant interactions were uncovered.

RESULTS

Animals: body weight and metabolic rate.

The body weight of 5-HT_1A KO pups was 17% lower than the body weight of WT pups at P5 weighing 3.4 ± 0.4 g versus 4.1 ± 0.5 g (mean ± SD; P < 0.001). The body weight, however, was not significantly different from WT pups at P15 (8.9 ± 1.2 g vs. 9.7 ± 2.5 g) and P25 (12.5 ± 02.4 g vs. 13.3 ± 2.4 g). The body weight of male and female pups of both genotypes was not significantly different. There were no apparent behavioral differences between WT and KO pups, though no formal behavioral testing was done. We did not observe any seizure-like activity. V̇o₂ was not different between KO and WT pups at any age (Fig. 2A).

Fig. 2.

Breathing in 5-HT_1A KO and WT pups in normoxia (RA) and during hypercapnia (5% CO₂) at P5, P15, and P25. A: KO pups had a significantly lower minute ventilation (V̇e) during normoxia and hypercapnia at P5 (*P < 0.01; interaction between gas and genotype) and P25 (*P < 0.05; interaction between gas and genotype). Oxygen consumption (V̇o₂) was not different between KO and WT pups at any age. V̇e/V̇o₂ was significantly lower in KO compared with WT pups at P5 (*P < 0.05; interaction between gas and genotype) and at P25 (*P < 0.01; effect of genotype). One-way repeated measures analysis of variance (ANOVA) at each age, with genotype as the between subjects factor and gas as the repeated factor; Bonferroni post hoc comparisons. B: The hypercapnic ventilatory response, expressed as the percent change in V̇e and V̇e/V̇o₂, was not different between KO and WT pups at any age (P > 0.05). One-way ANOVA at each age, with genotype as the between subjects factor. Data shown as means ± standard deviation (SD).

Breathing in normoxia and hypercapnia.

Minute ventilation (V̇e) during normoxia was significantly (16% and 27%) lower in KO compared with WT pups at P5 and P25, respectively (Fig. 2A; P < 0.001) but was not different between the two genotypes at P15 (Fig. 2A). The reduced V̇e was due to a slower f at P5 (P < 0.01) and a smaller V_T at P25 (P < 0.001; data not shown). During hypercapnia, breathing increased significantly in both KO and WT pups at all ages, but V̇e was still 23% and 31% lower in KO compared with W_T pups at P5 and P25, respectively (Fig. 2A; P < 0.0001). The hypercapnic ventilatory response (HCVR), expressed as the percent change in V̇e and V̇e/V̇o₂ from RA (Fig. 2B; P > 0.05), was not significantly different between KO and WT pups at any age. In summary, resting V̇e was significantly reduced in KO pups at P5 and P25, but the response to hypercapnia was intact at all ages.

Breathing variability.

Breathing variability, as measured by SD, RMSSD, and L × T of the Poincaré plot of the IBI, was 57%, 66%, and 187% higher in KO compared with WT pups, respectively, in normoxia at P5 (Fig. 3; P < 0.0001), and SD was 35% higher in P5 KO compared with WT pups during hypercapnia (P < 0.01). This increased breathing variability was, however, transient, such that, by P25, SD and RMSSD were 21% lower in KO compared with WT pups during normoxia (P < 0.05). At all ages and in both genotypes, hypercapnia caused a significant decrease in breathing variability (P < 0.001). There were no gender differences in breathing variability.

Fig. 3.

Interbreath interval variability in 5-HT_1A KO and WT pups in normoxia (RA) and during hypercapnia (5% CO₂) at P5, P15, and P25. Breathing variability, expressed as SD, root of the mean square of successive differences (RMSSD), and an estimate of the Poincaré distribution (L × T), was greater in KO compared with WT pups at P5 (*P < 0.05; interaction between gas and genotype), but was lower than WT pups at P25 (*P < 0.05; interaction between gas and genotype). One-way repeated measures ANOVA at each age, with genotype as the between subjects factor and gas as the repeated factor; Bonferroni post hoc comparisons. Data shown as means ± SD.

HR in normoxia and hypercapnia.

There were slight but significant alterations in HR in KO compared with WT pups. KO males and females had a 6% to 10% slower HR than WT pups independent of gas at P5 (P < 0.001), but, at P25, HR was 7% faster in KO males during normoxia (P < 0.001) and 9% faster in KO females during hypercapnia (P < 0.001). Figure 4, top, shows the data for both genders combined. Hypercapnia slowed HR in both WT and KO pups. In summary, HR in KO pups of both genders switched from being slower than controls at P5 to being faster than controls with increasing age.

Fig. 4.

HR and the low frequency to high frequency ratio (LF/HF) in 5-HT_1A KO and WT pups in normoxia (RA) and during hypercapnia (5% CO₂) at P5, P15, and P25.Top: HR was slower in KO compared with WT pups at P5 (***P < 0.0001; effect of genotype), but faster than WT pups at P25 (*P < 0.01; interaction between gas, gender, and genotype). Two-way repeated measures ANOVA at each age, with gender and genotype as the between subjects factors and gas as the repeated measure; Bonferroni post hoc comparisons. Bottom: LF/HF ratio was lower in KO compared with WT pups at P5 (*P < 0.05; effect of genotype) but higher than WT pups at P25 (*P < 0.05; effect of genotype). One-way repeated measures ANOVA at each age, with genotype as the between subjects factor and gas as the repeated factor. Data shown as means ± SD.

HRV.

The LF/HF ratio in KO pups was 47% lower than in WT pups at P5, but 68% greater than WT pups at P25 independent of gas (Fig. 4, bottom; P < 0.05). HRV, as measured by RMSSD and L × T, was ∼2-fold greater in KO compared with WT pups at P5 independent of gas (Fig. 5; P < 0.001 and P < 0.05, respectively) but was not different between the two genotypes at P15 and P25 (Fig. 5; P > 0.05). SD was also not different between the genotypes at any age (Fig. 5; P > 0.05).

Fig. 5.

Heart rate variability (HRV) in 5-HT_1A KO and WT pups in normoxia (RA) and during hypercapnia (5% CO₂) at P5, P15, and P25. HRV, expressed as RMSSD and L × T, was greater in KO compared with WT pups at P5 (*P < 0.05, **P < 0.001; effect of genotype), but was not different from WT pups at P15 and P25 (P > 0.05). One-way repeated measures ANOVA at each age, with genotype as the between subjects factors and gas as the repeated measure. Data shown as means ± SD.

Occurrence of spontaneous bradycardias.

When the bradycardia frequency was compared at each age, no differences were found between KO and WT pups at P5 or P15, either in normoxia or during hypercapnia. At P25 the bradycardia was ∼3-fold greater in KO compared with WT pups during hypercapnia (Fig. 6; P < 0.0001), and was 2-fold greater during hypercapnia compared with normoxia in KO pups only (Fig. 6; P < 0.001). The severity of the bradycardias, that is, the duration and magnitude of the HR change, was not different between KO and WT pups at any age, either in normoxia or during hypercapnia (data not shown). Also, there were no gender differences in either the bradycardia frequency or the severity of the bradycardias (data not shown).

Fig. 6.

Bradycardia frequency in 5-HT_1A KO and WT pups in normoxia (RA) and during hypercapnia (5% CO₂) at P5, P15, and P25. At P25, KO pups had more spontaneous bradycardias during hypercapnia compared with WT pups (***P < 0.0001; interaction between gas and genotype) and compared with normoxia (††P < 0.001; interaction between gas and genotype). One-way repeated measures ANOVA at each age, with genotype as the between subjects factor and gas as the repeated measure; Bonferroni post hoc comparisons. Data shown as means ± SD.

DISCUSSION

Main findings.

In this study, we used the 5-HT_1A receptor KO mouse to test the hypothesis that the 5-HT_1A receptor is important for development of cardiorespiratory control and that absence of this receptor results in alterations in resting ventilation and the ventilatory response to hypercapnia and impairs the normal development of autonomic control in early postnatal life. The main findings of the study are that, in KO pups at P5: 1) body weight is slightly but significantly reduced; 2) ventilation is reduced at baseline and during hypercapnia via a slowing of breathing frequency; 3) the HCVR is not affected; 4) breathing variability is increased at baseline; 5) HR is reduced; and 6) the LF/HF measure of HRV is reduced and RMSSD and L × T are increased. In KO pups at P25: 1) V̇e is reduced both at baseline and during hypercapnia; 2) the HCVR is not affected; 3) HR is increased at baseline and during hypercapnia; 4) the LF/HF measure of HRV is increased; and 5) the bradycardia frequency is increased during hypercapnia compared with baseline and compared with WT pups.

Characteristics of the 5-HT_1A receptor KO mouse.

Autoradiographic studies in the 5-HT_1A KO mouse show that the 5-HT_1A receptor is completely absent in the raphe nuclei of the brain stem, that is, the site of the 5-HT source neurons, and in their projection sites in the forebrain (24, 48, 55). In the absence of this receptor, adult mutant mice have an increase in the firing rate of 5-HT neurons (56); normal (3, 23, 55) or increased levels of 5-HT in the brain (49); and increased levels of the 5-HT metabolite, 5-hydroxyindolacetic acid (5-HIAA), and the 5-HIAA/5-HT ratio, the latter indicative of an increase in 5-HT turnover (3). In addition, these mice in adulthood display increased anxiety-like behavior (24, 48, 55). There are conflicting reports concerning the levels of 5-HT in the brain of 5-HT_1A mutant mice in early postnatal life (15, 30), but, as in adults, the levels of 5-HIAA and the 5-HIAA/5-HT ratio are increased (15), which are likewise indicative of an increase in 5-HT turnover, suggesting that 5-HT signaling is increased in 5-HT_1A mutant mice from early in development. In this study, we did not observe any apparent behavioral differences between 5-HT_1A KO and WT mice. The mutant mice were, however, slightly but significantly smaller than control mice at P5. 5-HT signaling has been linked to body weight control (19, 22), but the mechanism linking the 5-HT_1A KO mouse to reduced body weight at P5 is not clear. Apart from the slightly smaller size, 5-HT_1A KO pups developed normally, had normal metabolic rates, and did not display unusual postnatal mortality. We observed spontaneous mortality in 13% (3/22) KO mice and 30% (10/33) WT mice during the course of the studies. While we do not know the cause of this mortality, there was no statistically significant difference in mortality between KO and WT animals.

Breathing in normoxia and hypercapnia.

Baseline ventilation in 5-HT_1A KO pups was reduced at P5 and P25, which, given the unchanged metabolic rate, is indicative of hypoventilation in these mutant animals. Male 5-HT transporter (SERT) KO mice, which also develop in the presence of excess 5-HT in the brain, display a similar decrease in baseline ventilation in early postnatal life (52). We speculate that the hypoventilation observed in the 5-HT_1A mutants may be due to downregulation of excitatory postsynaptic 5-HT receptors in response to chronic activation from increased extracellular 5-HT levels (13), which could be present in both the 5-HT_1A and 5-HTT KO mice. Interestingly, animal models with a substantial decrease in 5-HT content also show a similar reduction in resting ventilation. For example, this occurs in tryptophan-deficient rat pups that have an ∼50% deficiency in medullary 5-HT (53), in neonatal TPH2 KO mice that are completely deficient in 5-HT (2), in neonatal Pet-1 null mice (20), and in 5,7-dihydroxytryptamine (5,7-DHT) lesioned rat pups that have a substantial loss of 5-HT neurons (14), and in neonatal Lmx1B null mice that have complete loss of 5-HT neurons (26). This suggests that respiratory output in early postnatal life is dependent on appropriate levels of 5-HT at the synapse and extracellular space.

One limitation of our study is that the whole body plethysmograph was used to assess ventilation at all postnatal ages. This can lead to an overestimation of the tidal volume and in turn an overestimation of ventilation, because an increase in the compression and rarefaction of air as it passes through the very narrow airways of these small animals can cause a greater pressure change within the plethysmograph than the pressure change associated with wetting and warming of inspired air (46). The use of whole body plethysmography, however, allows ventilatory measurements in conscious, unrestrained pups; and since this approach was used with both the KO and WT animals, it should not affect ventilatory comparisons between the two groups.

Hypercapnia did not have any further effect on the respiratory deficit in 5-HT_1A KO mice at any postnatal age. In fact, the HCVR, expressed as the percent change in V̇e and in V̇e/V̇o₂, was normal and appropriate for each developmental age studied, and appeared to follow the same age-dependent maturational trajectory as in WT mice. This came as a surprise since 5-HT neurons in the medullary raphe are known to contribute to central chemoreception, either by acting as central chemoreceptors themselves (25) or by modulating the activity of other central chemoreceptors (16, 17, 38). Furthermore, previous studies from our laboratory showed that the HCVR was enhanced in the first postnatal week and decreased at older ages after microdialysis of 8-OH-DPAT in the medullary raphe of newborn piglets (43) and is decreased in P15 and P25 rat pups from dams on a tryptophan-deficient diet (53). These data in the 5-HT_1A KO mice suggest that either the 5-HT_1A receptor is not involved in the 5-HT-mediated enhancement of CO₂ chemosensitivity in early postnatal life or that there are developmental compensations for the complete loss of both pre- and postsynaptic 5-HT_1A receptors that may be recruited during a hypercapnic challenge during the early postnatal period. In fact, the activity of the 5-HT_1B autoreceptor, which is also involved in regulating extracellular concentrations of 5-HT, has been found to be upregulated in the adult 5-HT_1A KO mouse (37).

Breathing variability.

The breathing pattern in newborn mammals is known to be highly irregular and unstable, particularly in the immediate postnatal period, but stabilizes with increasing age as the animal transitions to life ex utero (27, 45). Consistent with this, we found that, in both KO and WT pups, breathing variability was greatest during the first postnatal week and decreased with increasing age. KO pups, however, displayed transiently greater breathing variability than WT pups in the first postnatal week, which resolved with increasing age, suggesting an imbalance of excitatory and inhibitory modulation of respiratory network excitability. Several in vitro studies have demonstrated the importance of 5-HT acting through 5-HT_2A receptors within the respiratory network in generation and in stabilization of the respiratory rhythm (6, 7, 47, 51, 56, 59). Furthermore, in in vivo studies in Pet-1 null mice (20), 5,7-DHT lesioned rats (14) and Lmx1B null mice (26), which have a substantial or complete loss of 5-HT neurons, the breathing pattern is irregular and unstable in the neonatal period but stabilizes as the animals mature. This suggests that balanced 5-HT levels in the central nervous system are required for stabilization and timely maturation of the respiratory network and that the 5-HT_1A receptor plays a critical role in both of these processes.

HR during normoxia and hypercapnia.

While there are differences in absolute HR between KO and WT pups at P5 and P25, these effects are small. This suggests that there is redundancy in the neural control of cardiac function. The 5-HT7 receptor, for example, has been shown to excite cardiac preganglionic vagal neurons (CVPNs) via a mechanism similar to the 5-HT_1A receptor (12, 33). Furthermore, 5-HT2 and 5-HT3 receptors, which also have an excitatory effect on CVPNs, are expressed in the nucleus of the solitary tract (NTS; another brain stem nucleus involved in regulation of cardiac function) and in the nuclei where CVPNs reside (31, 54) and can therefore provide excitatory inputs to these CVPNs.

HRV.

Frequency domain analysis of HRV revealed that the LF/HF ratio in KO pups was significantly reduced at P5, but was significantly increased at P25. The LF/HF ratio, which is a reflection of the balance between sympathetic and parasympathetic inputs to cardiac control (41), suggests that, in the absence of the 5-HT_1A receptor, parasympathetic regulation of cardiac function is increased at the younger age, but is decreased at the older age. According to Longin et al. (39), maturation of the autonomic nervous system is accompanied by increasing HRV with a pronounced increase in parasympathetic activity (11, 21, 39, 57). The LF/HF ratio seems to decrease with age in WT pups but remains constant across all ages in the KO pups, which would suggest a delay in the maturation of autonomic processes controlling cardiac variability in these mutant animals. Activation of 5-HT_1A receptors in astroglial cells has been shown to be important for maturation of neurons (5, 29), and the subtle delay in maturation of the autonomic nervous system in the KO mice shown in this study suggests that the 5-HT_1A receptors may be important for maturation of the autonomic nervous system function as well.

Occurrence of spontaneous bradycardias.

At baseline, the frequency of spontaneous bradycardias was not different between KO and WT pups at any postnatal age studied. When challenged with hypercapnia, the bradycardia frequency in KO pups was higher than in WT pups and higher than at baseline; however, these differences were only significant at P25. It is not clear why KO pups have a greater bradycardia frequency during hypercapnia at P25, especially since the LF/HF ratio suggests that vagal tone is less in KO pups at this age than in WT pups. Whether this is due to an alteration in autonomic control that is only apparent after a homeostatic challenge is uncertain. These data, however, suggest that, in the absence of the 5-HT_1A receptor, the decrease in inhibitory inputs to the CVPNs, that is induced by hypercapnia, is exaggerated at older ages.

Conclusions.

Despite complete absence of 5-HT_1A receptors, the effects on cardiorespiratory control during early postnatal development are moderate, subtle, and primarily focused at P5 and thus likely reflect the immediate transition period from intrauterine to extrauterine life. In 5-HT_1A KO pups compared with WT at P5, body weight was less, ventilation was reduced due to a slower and more variable breathing frequency, and HR was slower and also more variable. Thus there is evidence of cardiorespiratory instability at P5 that then resolves with further postnatal development. A primary impetus for these studies was the reproducible finding in many SIDS cases of reduced 5-HT_1A receptor binding in the medullary nuclei that contain 5-HT source neurons and those that receive projections (see Introduction). These mice display an exaggeration of one feature of the described constellation of abnormalities involving several neurotransmitter systems and the 14-3-3 signal transduction family described to date in SIDS brain stems (8, 9, 35). Thus, the 5-HT_1A KO mouse is an incomplete animal model for neurochemical brain stem pathology in affected SIDS infants. Yet, the 5-HT_1A KO mouse demonstrates a critical age-dependent period of dysfunctional control of breathing and HR that may indicate a role for this receptor in the normal development of these critical processes and that may be relevant to elucidating its role in brain stem abnormalities and putative abnormal cardiorespiratory homeostasis in SIDS infants in a critical time frame. Moreover, this mutant may serve as a model for determining the mechanisms whereby an immutable genetic deficiency in the number of 5-HT_1A receptors is in some way compensated, such that the absolute deficiency results in a critical period of potential vulnerability at P5 when pups hypoventilate and have a slower respiratory frequency and HR with enhanced variability of both, suggesting abnormal maturation of cardiorespiratory control.

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

Author contributions: K.T.B., A.L., J.A.D., and E.E.N. conception and design of research; K.T.B. performed experiments; K.T.B. and J.C.L. analyzed data; K.T.B. prepared figures; K.T.B. drafted manuscript; K.T.B., H.C.K., A.L., J.A.D., J.C.L., and E.E.N. approved final version of manuscript; H.C.K., A.L., J.A.D., and E.E.N. interpreted results of experiments; H.C.K., J.C.L., and E.E.N. edited and revised the manuscript.