Brainstem serotonin deficiency in the neonatal period: autonomic dysregulation during mild cold stress

Abstract

Based on previous studies in adult animals, devoid of 5-HT neurones, showing altered thermoregulation in cold stress (4°C) and a reduced ventilatory response to CO₂, we hypothesized that neonatal mice lacking 60-70% of their 5-HT neurones (Pet-1(-/-)) would have: (1) a reduced thermogenic response to a mild drop in ambient temperature (TA), (2) reduced V(E) and heart rate (HR) responses to mild cooling that reflect this reduced thermogenic response, and (3) a reduced ventilatory response to CO₂ after postnatal day 12 (P12), when 5-HT neurones become chemosensitive in vitro. We first determined that a 60-70% loss of 5-HT-positive neurones results in a ~90% loss of 5-HT from the brainstems of Pet-1(-/-) animals. We then subjected Pet-1(-/-) and wild-type (WT) mice (N = 5) to mild environmental cooling (T(A) = 29°C) at ~P12. T(A) was initially held at 34°C for ~20 min, reduced to 29°C over 15 min and held for an additional 10 min at steady state, and then returned to 34°C. From 34°C to 29°C, there was a robust increase in V(O₂) in P12WT, but not Pet-1(-/-) animals (68±19.9% versus -16±8%, respectively; P = 0.002). On average, body temperature (T(B)) dropped 1.1°C more in Pet-1(-/-) compared to WT animals (P = 0.03). HR remained unchanged in WT but dropped 22±2.3% in Pet-1(-/-) animals (P = 0.01). Genotype had no effect on tail temperature (T(T)), either at 34°C or 29°C. After cooling, values for V(O₂) and HR of Pet-1(-/-) animals were no different to values predicted by Q₁₀ effects alone, while values of WT animals were greater than predicted. V(E) increased in WT with cooling, while it decreased in Pet-1(-/-) animals (P = 0.002). Still, Pet-1(-/-) animals hyperventilated relative to WT (increased V(E)/V(O₂)) irrespective of T(A) (P = 0.002). As tested in a separate group of pups, there was no difference in the ventilatory response to CO₂ between WT and Pet-1(-/-) animals, either at P5 or P15. We conclude that during neonatal life in mouse pups: (1) brainstem 5-HT is critical for the thermogenic response to a mild drop in environmental temperature probably via a sympathetically-mediated increase in brown fat metabolism; (2) reduced thermogenesis probably contributes to the reduced HR and V(O₂) observed with 5-HT deficiency; and (3) the presence of some brainstem 5-HT is sufficient for an appropriate ventilatory response to hypercapnia up until P15. Infants with reduced brainstem 5-HT could be prone to cardiovascular and respiratory abnormalities resulting from compromised thermogenesis.

Figure 2. Reduced thermogenesis and heart rate (HR) in Pet-1^−/− (KO) neonates during ambient cooling

A, raw traces showing the changes in the fractional concentration of O₂ leaving the mask (mask O₂) and body temperature (T_B) in Pet-1^−/− animals (n = 5, grey lines) and littermates (WT) (n = 5, black lines) as ambient temperature (T_A) is cooled from 34°C to 29°C over 25 min (shaded region). B, at 34°C, metabolic rate is not significantly different in Pet-1^−/− animals compared to WT. The disparity between genotypes becomes apparent as T_A is reduced to 29°C (shaded region). C, at 34°C, T_B is slightly but significantly lower in Pet-1^−/− animals compared to WT. As is the case with , the disparity becomes greater with cooling to 29°C. D, there were no differences in tail temperature (T_T) between Pet-1^−/− and WT animals at either T_A. Note: T_T is less than 29°C because of a T_A gradient within the chamber (T_A was 1.5°C lower at the tail end compared to the middle). E, there is a significantly greater reduction in HR with cooling in Pet-1^−/− animals compared to WT. All data in B–E are means ± SE. *Significant genotype effect, P < 0.05; †genotype ×T_A interaction, P < 0.05.

Figure 1. Pet-1^−/− animals (KO) have reduced brainstem 5-HT

Shown is the 5-HT, 5-hydroxyindoleacetic acid (5-HIAA) and noradrenaline (NA) content from the medulla and pons of wild-type WT animals (n = 4) and Pet-1^−/− animals (n = 4) at P7.

Figure 3. Effects of body size and Q₁₀ on thermogenesis and HR during ambient cooling

A and B, WT animals (filled circles, nos. 1–5) weigh more than Pet-1^−/− animals (open circles, nos. 6–10), and mass has a significant relationship with T_B and (2-factor repeated-measures ANOVA: P < 0.05 for each). C and D, and HR after cooling (normalised to baseline values) for WT (filled bars), and Pet-1^−/− (open bars). Shown are observed values and values predicted from the Q₁₀ effect alone (assuming Q₁₀ co-efficient of 2 and with a drop in T_B of 4.2°C in Pet-1^−/− and 3.1°C in WT). Data in C and D are means ± SE. *Significant difference exists between observed WT values for and HR and the values predicted from the Q₁₀ effect alone. No such difference exists between the observed and predicted values for Pet-1^−/− animals.

Figure 4. Hyperventilation in P12 Pet-1^−/− animals

A, is elevated in P12 Pet-1^−/− animals, irrespective of T_A. B, from T_A= 34°C to 29°C, decreases in Pet-1^−/− animals, but increases in WT animals. C and D, is selectively reduced in Pet-1^−/− animals from a reduction in tidal volume (V_T) and respiratory frequency (f) E, respiratory variability, as measured by the co-efficient of variation in the respiratory period (CV%) is higher in Pet-1^−/− animals at both T_A. All data are means ± SE. *Significant genotype effect, P < 0.05; †genotype ×T_A interaction, P < 0.05.

Figure 5. A loss of brainstem 5-HT neurones has no effect on the ventilatory response to CO₂ at P5 or P15

A, ventilation of WT and Pet-1^−/− (KO) animals at P5 and P15, during room air and after 2 min of 5% CO₂ balanced with air (CO₂–air). B, of WT and Pet-1^−/− animals during room air and after 2 min of 5% CO₂ balanced with O₂ (CO₂–O₂). C, increase in (% change) after 2 min of either CO₂–air or CO₂–O₂ in each genotype at P5 and P15. All data are means ± SE.