Fig. 2

Integrated two-photon imaging and physiological monitoring reveal the respiratory and heartbeat origins of vasomotion in anesthetized mice. (A) Schematic of the integrated experimental setup combining two-photon imaging, respiratory monitoring, and heartbeat recording in an anesthetized mouse. (B) Representative still-frame and kymographs of a pial arteriole showing vasomotion cycles synchronized with heartbeat activity. (C) The correspondence in time-series curves of arteriolar radius vasomotion with the heartbeat cycles. (D) The cardiac oscillatory component was isolated, yielding a markedly higher CC value between vasomotion and ECG after applying a heartbeat-centered bandpass filter. (E) Group analysis of cardiac-band cross-correlation between vasomotion and ECG (n = 18 pairs in 3 mice). (F) The representative Fourier plot across of arterioles radius change and heartbeat monitor revealed a broad range of frequencies (0–6 Hz), with a distinct peak centered at 3.722 Hz (heartbeat recorded in two-photon microscopy vasomotion) and 3.798 Hz (heartbeat recorded in heartbeat monitor). (G) Comparison of vasomotion frequencies, driven by heartbeat activity (left), and heartbeat frequencies measured by a heartbeat monitor (right) in three anesthetized mice. (H) Photograph of the imaging setup during video recording, the selected region of interest (ROI), highlighted on the mouse’s scalp, was chosen to capture breathing-induced motion for respiratory frequency extraction. (I) Motion energy plot derived from the ROI shown in (H), representing respiratory cycles (red), peaks labeled 1–14 correspond to distinct breathing events, identified based on fluctuations in motion energy associated with respiratory movements. (J) Time-series traces of arteriole diameter (red) and radius (magenta) recorded during 14 respiratory cycles, while the radius trace reveals oscillations corresponding to the respiratory frequency band identified in (I), no clear respiratory vasomotion frequency are observed in diameter trace. (K) The representative Fourier plot across of arterioles diameter and radius change and respiratory detected video revealed a range of frequencies (0.3–1.5 Hz), a distinct respiratory peak was identified at ~ 0.926 Hz in the respiration trace, corresponding to both diameter and radius vasomotion. The respiratory frequency band was more prominently represented in the radius spectrum, while the signal in the diameter trace was weaker and potentially confounded with other low-frequency components. (L) Respiratory and radius-based vasomotion frequencies both declined over time under prolonged isoflurane anesthesia, showing strong temporal concordance. The gray column indicates the induction phase (3% isoflurane, 0–3 min), after which 1.5% isoflurane was used for maintenance. The first time point for each mouse marks the start of simultaneous video recording and two-photon imaging. Statistics for (G), paired t-tests were used.