Elevated blood pressure, heart rate and body temperature in mice lacking the XLαs protein of the Gnas locus is due to increased sympathetic tone

Abstract

Imbalances of energy homeostasis are often associated with cardiovascular complications. Previous work has shown that Gnasxl-deficient mice have a lean and hypermetabolic phenotype, with increased sympathetic stimulation of adipose tissue. The Gnasxl transcript from the imprinted Gnas locus encodes the trimeric G-protein subunit XLαs, which is expressed in brain regions that regulate energy homeostasis and sympathetic nervous system (SNS) activity. To determine whether Gnasxl knock-out (KO) mice display additional SNS-related phenotypes, we have now investigated the cardiovascular system. The Gnasxl KO mice were ∼20 mmHg hypertensive in comparison to wild-type (WT) littermates (P ≤ 0.05) and hypersensitive to the sympatholytic drug reserpine. Using telemetry, we detected an increased waking heart rate in conscious KOs (630 ± 10 versus 584 ± 12 beats min(-1), KO versus WT, P ≤ 0.05). Body temperature was also elevated (38.1 ± 0.3 versus 36.9 ± 0.4°C, KO versus WT, P ≤ 0.05). To investigate autonomic nervous system influences, we used heart rate variability analyses. We empirically defined frequency power bands using atropine and reserpine and verified high-frequency (HF) power and low-frequency (LF) LF/HF power ratio to be indicators of parasympathetic and sympathetic activity, respectively. The LF/HF power ratio was greater in KOs and more sensitive to reserpine than in WTs, consistent with elevated SNS activity. In contrast, atropine and exendin-4, a centrally acting agonist of the glucagon-like peptide-1 receptor, which influences cardiovascular physiology and metabolism, reduced HF power equally in both genotypes. This was associated with a greater increase in heart rate in KOs. Mild stress had a blunted effect on the LF/HF ratio in KOs consistent with elevated basal sympathetic activity. We conclude that XLαs is required for the inhibition of sympathetic outflow towards cardiovascular and metabolically relevant tissues.

Figure 2. Circadian heart rate (HR), activity and HR responses in conscious Gnasxl KO mice

Heart rate and activity were recorded continuously in freely moving KO mice and WT siblings via ECG telemetry. A, 1 h averaged circadian HR over 24 h. Significantly different time points are indicated (*P≤ 0.05, n= 12, repeated-measures ANOVA). Filled bar, night period; open bar, day period. B, mean HR is increased at night in KO compared with WT mice (*P≤ 0.05), but did not reach significance during the daytime. C, 1 h averaged circadian activity over 24 h. D, mean activity is unchanged in KO compared with WT mice. E–G, treatments and HR recordings were undertaken during the light period. E, following i.p. injection of 1 mg kg⁻¹ of the sympatholytic reserpine the HR was significantly decreased in KO mice and trended to be lower in WT mice. F, HR was significantly increased in KO mice following i.p. injection of 2 mg kg⁻¹ of the parasympatholytic atropine. G, HR was significantly increased, to a similar extent, in both genotypes following mild handling stress. *P≤ 0.05 by Student’s paired t test. Error bars indicate SEM.

Figure 1. Blood pressure (BP), reserpine response and body temperature in anaesthetized Gnasxl knock-out (KO) mice

A, mean basal BP (MAP) measured via tail volume pressure recording plethysmography under light general anaesthesia and thermoneutral conditions was significantly increased in KO mice compared with wild-type (WT) mice (*P≤ 0.05, n= 9 and 8). The systolic and diastolic data for these mice are as follows: systolic, WT 76.7 ± 5.1 mmHg versus KO 102.6 ± 6.4 mmHg, P= 0.0072; and diastolic, WT 58.3 ± 4.2 mmHg versus KO 74.9 ± 6.4 mmHg, P= 0.053. B, the BP change in response to reserpine was significantly greater in KO compared with WT mice, relative to vehicle, as measured via arterial cannulation (*P≤ 0.05, n= 5). C, KO mice had significantly elevated body temperature as measured via rectal probe (*P≤ 0.05, n= 7). Error bars indicate SEM.

Figure 3. Heart rate variability (HRV) responses

Treatments and HR recordings by telemetry in conscious Gnasxl KOs and WT siblings were carried out during the light period. Low-frequency/high-frequency (LF/HF) ratio was used as an indicator of sympathetic stimulation of the cardiovascular system, specifically of sympathovagal balance, and HF power as an indicator of parasympathetic stimulation. A, WT and KO mice had a significant decrease in LF/HF ratio following i.p. injection of 1 mg kg⁻¹ reserpine. B, KO mice had a significantly greater relative decrease in LF/HF ratio compared with WT mice. C, both genotypes had a significant decrease in HF power following i.p. injection of 2 mg kg⁻¹ atropine. D, both genotypes had a similar relative decrease in HF power. E, LF/HF ratio was significantly increased in WT mice following stress, but not in KO mice. F, the comparison between genotypes of relative LF/HF change following stress did not reach significance. *P≤ 0.05 by Student’s paired t test, or by Student’s unpaired t test in B. Error bars indicate SEM.

Figure 4. Basic HRV in conscious Gnasxl KO mice

Heart rate spectra were produced as in Fig. 3. A, typical HR spectra for WT (Aa) and KO mice (Ab). B, LF/HF ratio was calculated for a minimum of 10 HR spectra for each mouse during the active dark period; mean LF/HF ratio was increased in KO mice. *P≤ 0.05, n= 9 and 12 for KO and WT, respectively. Error bars indicate SEM.

Figure 5. Heart rate and HRV responses to exendin-4 (Ex-4)

Treatments and HR recordings were carried out during the light period as in Figs 2 and 3. A, HR response over time in response to i.p. injection of 50 μg kg⁻¹ Ex-4 (arrow). B, HR is significantly increased in KO mice following Ex-4 and trended higher in WT mice. **P≤ 0.01 by Student’s paired t test. C and D, there was no significant difference in LF/HF ratio or relative change in LF/HF ratio following Ex-4 in either genotype. E and F, both genotypes had a decrease in HF power following Ex-4, although this reached significance only in WTs. The relative decrease in HF power was significant to a similar extent. *P≤ 0.05 and ***P≤ 0.001 by Student’s paired t test. Error bars indicate SEM.

Figure 6. Neuronal c-fos response to Ex-4 in the hypothalamic paraventricular nucleus (PVN)

c-fos immunohistochemistry of coronal brain sections after injection of 50 μg kg⁻¹ i.p. Ex-4. A, representative images showing high levels of c-fos in the PVN of both genotypes. B, no significant c-fos response was observed following saline injection. C, there was no significant difference in numbers of c-fos-positive neurones (per section and left/right brain side) between genotypes. This was also confirmed in a more detailed anterior–posterior PVN subregion-specific assessment (data not shown). Error bars indicate SEM.

Figure 7. c-fos response to Ex-4 in correlation to XLαs expression in the PVN

Immunofluorescence for c-fos (green) and XLαs (red) on brain sections of WT mice after Ex-4 i.p. injection. A representative image demonstrates XLαs-expressing neurones located around the periphery of the PVN, while all c-fos positive neurones are found in the centre of the PVN in a noticeably separate cell population. Of five WT mice investigated, co-localization between XLαs and c-fos was <1%.