Dim light at night worsens cardiac autonomic dysregulation in female diabetic mice

Cardiovascular autonomic neuropathy (CAN) is a common complication of diabetes mellitus. CAN is characterized by impaired cardiovascular autonomic control and is associated with increased mortality, especially among women with type 2 diabetes.¹ Identifying environmental factors that exacerbate the CAN-associated phenotype may reveal modifiable strategies to reduce cardiovascular risk.

Nighttime light exposure is a novel cardiovascular risk factor.² Among ~90,000 individuals, greater nighttime light exposure correlated with increased atrial fibrillation and heart failure risk.³ Light at night disrupts sympathovagal balance and elevates nocturnal heart rate in humans.² Dim light at night similarly disrupts autonomic regulation of heart rate in mice.⁴ We hypothesized that dim light at night would exacerbate CAN-associated phenotypes in female diabetic mice. We found that (1) thermoneutral housing unmasked diabetic cardiac phenotypes, (2) dim light at night worsened CAN-associated tachycardia and abolished circadian variation in heart rate, and (3) an abnormally prolonged temporal lag between correlated heart rate and core body temperature fluctuations was a consistent feature of the CAN-associated phenotype.

Female db/db mice (Lepr^−/−; JAX stock no. 000642), a type 2 diabetes model exhibiting obesity, hyperglycemia, and hypoactivity, and control mice (Lepr^+/+), aged 2–3 months, were implanted with PhysioTel ETA-F10 telemetry transmitters (Data Sciences International). After a 2-week recovery under 12-hour light/12-hour dark cycles, heart rate (RR interval) and core body temperature were continuously recorded using Ponemah (version 6.42; Data Sciences International) and analyzed with Prism (version 10.6.1; GraphPad). Procedures followed the Association for Assessment and Accreditation of Laboratory Animal Care guidelines with University of Kentucky Institutional Animal Care and Use Committee approval.

Revealing CAN-associated tachycardia required thermoneutral housing (30°C)

Standard laboratory room temperatures (22°C–25°C) fall below the thermoneutral zone in mice, necessitating increased sympathetic tone for thermoregulation, which may mask pathological autonomic changes. Figure 1A shows hourly heart rate and core body temperature in control (n = 6; top graph) and db/db (n = 5; bottom graph) female mice housed sequentially at room temperature (25°C ± 1°C, 3 days) and thermoneutrality (30°C ± 1°C, 5 days) under 12-hour light (200 lux)/12-hour dark (0 lux) cycles. Mice are nocturnal, and their heart rate and core body temperature peak during the dark cycle. At room temperature, db/db mice maintained heart rates comparable to control mice despite lower core body temperatures (Figure 1B and 1C). Thermoneutral housing normalized temperature differences and unmasked tachycardia in db/db mice during both light and dark cycles.

Figure 1.

Thermoneutral housing unmasks diabetic cardiac phenotypes, reveals light at night as an environmental stressor that exacerbates cardiovascular autonomic neuropathy–associated dysfunction, and identifies prolonged heart rate-temperature lag as a potential index of an autonomic impairment in diabetes. A: The hourly mean (standard deviation [SD]) heart rate (HR; red circle) and core body temperature (Tb; light blue circle) data were plotted for control and db/db mice under room temperature (RT), thermoneutrality (TN), and dim light at night (dLAN) conditions. B: Average day (12-hour light) and average night hours (12-hour dark/dim light) HR and Tb in control and db/db mice. C: The histogram shows the number of control and db/db mice exhibiting significant (filled) and nonsignificant (empty) 24-hour rhythms in HR and Tb under 3 conditions (RT, TN, and dLAN), as determined by cosinor analysis. D: The cross-correlation relationship between 10-second fluctuations in HR and Tb in control (top) and db/db (bottom) mice. E: Mean (SD) lag (in minutes) between HR and Tb in control and db/db mice housed under RT, TN, and dLAN conditions. Data in panels B and E were analyzed using a 2-way repeated-measures analysis of variance followed by the Tukey post hoc test. Normality was assessed using the Shapiro-Wilk test. P < .05 was considered statistically significant. LD = 12-hour light/12-hour dark.

Dim light at night worsened CAN-associated tachycardia and abolished circadian heart rate variation

To test whether dim light at night exacerbated the cardiac CAN-associated phenotype, we exposed mice housed at thermoneutrality to a 12-hour light (200 lux)/12-hour dim light (5 lux) cycle for 8 days. Dim light at night exposure increased daytime heart rates in db/db mice (Figure 1B), disrupting circadian variation in heart rate. No increased daytime activity was observed in db/db mice, suggesting that faster daytime heart rates were not secondary to changes in locomotion (data not shown).

Cosinor analysis assessed the circadian rhythmicity of heart rate and core body temperature in individual mice.⁴ Most control mice exhibited significant circadian rhythms across all conditions (Figure 1C). Most db/db mice exhibited circadian rhythms at room temperature and thermoneutrality; however, dim light at night abolished the circadian rhythm in heart rate while preserving core body temperature rhythms. This selective disruption shows that dim light at night differentially affected the integrated circadian regulation of cardiovascular vs thermoregulatory outputs in diabetic mice.

Abnormal integrated short-term regulation between heart rate and core body temperature fluctuations is a stable feature of CAN

Short-term coupling between rapid (10-second) heart rate and core body temperature fluctuations was analyzed using cross-correlation on 48-hour data sets after locally estimated scatterplot smoothing detrending to remove circadian components. Maximal cross-correlations (r = 0.4–0.6) indicated moderate time-shifted linear covariation (Figure 1D). In control mice, heart rate changes preceded correlated changes in core body temperature by 4–6 minutes (Figure 1E). Conversely, db/db mice exhibited prolonged lags (9–14 minutes) across all conditions, suggesting delayed coupling is a stable feature of CAN.

One limitation is the absence of time-matched db/db control mice to isolate long-term effects of thermoneutral housing. Our experimental design emphasized within-mouse comparisons to test for changes in CAN-associated phenotypes. Our control and prior longer-term studies show that circadian heart rate and temperature rhythms are preserved in mice housed at 30°C, supporting the interpretation that loss of rhythmicity in db/db mice likely reflects diabetic autonomic dysfunction interacting with dim light at night.⁴

Conclusion

Thermoneutral housing unmasked tachycardia in db/db mice consistent with human CAN. Dim light at night further elevated daytime heart rate and abolished circadian heart rate variation in female diabetic mice. Across conditions, db/db mice exhibit delayed temporal coupling between heart rate and temperature. These findings show the importance of thermoneutral housing for revealing diabetic cardiac phenotypes, identify nighttime light as an environmental stressor that exacerbates CAN-associated cardiac dysfunction, and suggest that prolonged heart rate–temperature lag may identify an autonomic impairment in diabetes. Our results support minimizing nighttime light exposure as a potentially modifiable intervention to reduce cardiovascular risk in diabetic patients with CAN.