Mapping cerebral perfusion in mice under various anesthesia levels using highly sensitive BOLD MRI with transient hypoxia

Thuy Thi Le; Geun Ho Im; Chan Hee Lee; Sang Han Choi; Seong-Gi Kim

doi:10.1126/sciadv.adm7605

. 2024 Feb 28;10(9):eadm7605. doi: 10.1126/sciadv.adm7605

Mapping cerebral perfusion in mice under various anesthesia levels using highly sensitive BOLD MRI with transient hypoxia

Thuy Thi Le ^1,^2,³, Geun Ho Im ¹, Chan Hee Lee ¹, Sang Han Choi ¹, Seong-Gi Kim ^1,^2,^3,^*

PMCID: PMC10901365 PMID: 38416820

Abstract

Cerebral perfusion is critical for the early detection of neurological diseases and for effectively monitoring disease progression and treatment responses. Mouse models are widely used in brain research, often under anesthesia, which can affect vascular physiology. However, the impact of anesthesia on regional cerebral blood volume and flow in mice has not been thoroughly investigated. In this study, we have developed a whole-brain perfusion MRI approach by using a 5-second nitrogen gas stimulus under inhalational anesthetics to induce transient BOLD dynamic susceptibility contrast (DSC). This method proved to be highly sensitive, repeatable within each imaging session, and across four weekly sessions. Relative cerebral blood volumes measured by BOLD DSC agree well with those by contrast agents. Quantitative cerebral blood volume and flow metrics were successfully measured in mice under dexmedetomidine and various isoflurane doses using both total vasculature-sensitive gradient-echo and microvasculature-sensitive spin-echo BOLD MRI. Dexmedetomidine reduces cerebral perfusion, while isoflurane increases cerebral perfusion in a dose-dependent manner.

Whole-brain perfusion mapping in mice under various anesthetic levels using BOLD-MRI is demonstrated.

INTRODUCTION

Cerebral perfusion plays a crucial role in maintaining homeostasis by delivering oxygen and nutrients while eliminating wastes. It serves as a valuable biomarker for various disorders, including ischemic stroke, dementia, Alzheimer’s disease, tumors, and others (1). Thus, noninvasive measurements of perfusion metrics such as cerebral blood volume (CBV) and cerebral blood flow (CBF) are essential for effectively monitoring disease progression and treatment response. To measure CBV and CBF noninvasively, magnetic resonance imaging (MRI), positron emission tomography (PET), and dynamic computed tomography (CT) are widely used with exogenous tracers (2). However, repeated measurements of perfusion metrics are limited due to concerns of excessive exposure to ionizing radiation in PET and CT, as well as exceeding the allowed dose for gadolinium (Gd) contrast agents in MRI for human subjects. These problems may be solved by using endogenous paramagnetic deoxyhemoglobins (dHb) as a safe blood pool contrast (3–5) in conjunction with blood oxygenation level dependent (BOLD) dynamic susceptibility contrast (DSC) MRI, which requires the precise and brief delivery of transient hypoxia. However, application of BOLD DSC to animals under inhaled anesthesia is limited due to the difficulty of controlling hypoxic stimuli precisely without interference from the anesthesia itself.

To investigate neurovascular diseases, mouse models are commonly used in biomedical research, since they allow for invasive procedures, genetic manipulation, and testing of pharmacological interventions (6–9). Anesthesia is routinely used in mouse experiments, which affects vascular physiology. Isoflurane, a commonly used anesthetic in animal research (10, 11), causes dose-dependent vasodilation (12, 13). On the other hand, dexmedetomidine, often used for functional studies (14–17), acts as a vasoconstrictor (15, 18). The effect of anesthesia on regional CBF in a few selected slices of mice has been investigated with arterial spin labeling (ASL) MRI (19–21), but its impact to regional CBF and CBV in the whole brain has not been reported in mice. Therefore, it is crucial to determine baseline perfusion values under commonly used anesthetics for understanding the regional effects of anesthesia and for accurately interpreting experimental results in mouse models.

In the present work, we developed a hypoxia-induced BOLD-DSC MRI setup for mapping perfusion in mice under volatile anesthesia. We assessed the sensitivity and repeatability of BOLD responses within a single imaging session and across multiple sessions and compared CBV measured by BOLD-DSC and conventional contrast agent MRI. Then, we investigated the effects of different doses of isoflurane (ISO) on whole-brain and regional CBV and CBF, comparing these results with those obtained under dexmedetomidine/isoflurane (DEX + ISO). Our findings demonstrated that the BOLD-DSC MRI response to a 5-s anoxic challenge is highly repeatable within each imaging session and across imaging sessions within a month, and relative CBV values measured by BOLD-DSC and contrast agent MRI agree extremely well. By combining gradient-echo (GE) and spin-echo (SE) BOLD measurements, we were able to estimate the composition of macrovascular and microvascular components on a regional basis. We found that regional perfusion values were lowest under DEX + ISO anesthesia and increased with higher doses of ISO.

RESULTS

Experimental setup for BOLD-DSC perfusion measurements under volatile anesthesia

Tissue perfusion can be measured by observing the response of capillaries and nearby tissue to the modulation of arterial blood signals. In dynamic BOLD-DSC studies (3–5), it is crucial to have a precise gas challenge for obtaining reliable and repeatable measurements. In our previous studies (4), we adopted a single gas delivery line for the animal, causing delays in nitrogen gas delivery and limitations in combining volatile anesthetics. To address these issues, we modified our original setup by incorporating two separate gas lines, one for medical gas and the other for nitrogen (N₂) gas, which ensures fast and accurate hypoxic gas challenges, even under inhalational anesthesia (Fig. 1A).

To effectively average BOLD-DSC responses to hypoxic stimuli, a block-design paradigm of 5-s anoxia and 60-s rest was repeated five times (Fig. 1B and see Materials and Methods). The 5-s anoxic stimulus reduced an arterial oxygen saturation level to 59% (4) and consistently induced similar GE-BOLD responses with peak intensities of approximately −11% in the primary somatosensory cortex (SS; Fig. 1, C and F; −11.2 ± 1.5%). Since ISO was not administered during the short stimulus period, the GE-BOLD responses may reflect anesthesia-modulated physiological changes. To address this concern, we conducted BOLD-DSC studies with a normoxic gas challenge (see Materials and Methods) and found no substantial change in the GE-BOLD responses (Fig. 1, D and F; 0.03 ± 0.2%), indicating that the absence of volatile anesthesia during the short stimulation period (5 s of 65 s) did not induce observable vascular variations. When a 5-s apnea was used instead of anoxia (see Materials and Methods), a significantly smaller-negative BOLD response was detected (Fig. 1, E and F; −1.07 ± 0.12%). However, when a longer apnea stimulus was used, it resulted in increased negative BOLD responses (see fig. S1). On the basis of these findings, our hypoxic stimulation setup is deemed suitable for perfusion mapping under volatile anesthesia.

Optimization of echo time for BOLD-DSC MRI

In DSC MRI, it is critical to maximize hypoxia-induced signal changes and to accurately determine arterial input functions. Thus, optimizing an appropriate echo time (TE) is vital, since it influences both the baseline signal-to-noise ratio (SNR) and the magnitude of hypoxia-induced changes. A shorter TE results in reduced hypoxia-induced changes, while a longer TE decreases the baseline SNR and increases the risk of arterial signal saturation (22, 23). Vascular voxels are susceptible to peak saturations when the raw MR signal approaches the background noise level during the first pass of the contrast agents (24) (see Materials and Method section). Thus, we varied TE between 11.6 and 20 ms in six mice and measured hypoxia-induced time curves and $∆ R_{2}^{*}$ of arterial, venous, and somatosensory tissue voxels.

The effects of longer TE on baseline signal and SNR were clearly observed in artery, vein, and tissue voxels (Fig. 2, A to D). Baseline $T_{2}^{*}$ values were 24.07 ± 2.17 ms for the artery, 9.23 ± 0.59 ms for the vein, and 23.81 ± 3.00 ms for the tissue (Fig. 2H), which are consistent with previous findings, where tissue baseline $T_{2}^{*}$ was 26.6 ms (25). Although the optimal TE for maximizing contrast is close to $T_{2}^{*}$ value, it is essential to ensure sufficient sensitivity to capture the hypoxia-induced peak for accurate quantification of relaxation rate changes ( $∆ R_{2}^{*}$ ).

We observed peak saturations in the arterial voxel at TEs of 15, 17, and 20 ms (Fig. 2A, expanded plot) and in the venous voxel at TEs of 17 and 20 ms (Fig. 2B). The signal time curves of the artery, vein, and tissue were then converted to concentration time curves ( $∆ R_{2}^{*}$ ) (see Materials and Methods). The noisy signal time curves of the artery and vein acquired at longer TEs of 15, 17, and 20 ms resulted in lower concentration time curves (Fig. 2, E and F). This indicates that longer TEs lead to peak saturations and result in underestimation of the arterial and venous concentration time curves. However, at shorter TE values of 11.6 and 13 ms, significantly higher and similar concentration time curves were induced in the artery and vein. In contrast, no notable differences were observed in the $∆ R_{2}^{*}$ curves obtained at all different TEs in the tissue (Fig. 2G). On the basis of these findings, we have selected the shortest achievable TE value for further studies.

Sensitivity and repeatability of hypoxia-induced BOLD MRI within a single imaging session

Since high reliability is critical for routine perfusion MRI studies, we evaluated the repeatability and sensitivity of GE BOLD-DSC experiments in 10 mice under 1.5% ISO anesthesia. The BOLD-DSC MRI run consisted of five 5-s hypoxic trials was repeated three times, resulting in a total of 15 hypoxic trials (Fig. 3A).

Fig. 3. — (A) Hypoxic stimulation paradigm. Three experimental runs were conducted, with each run including 5 hypoxic trials, resulting in a total of 15 trials for one session. (B) Single-slice signal change (∆S) maps across 15 hypoxic trials of one ISO-anesthetized mouse. ∆S maps were shown in only 9 trials of the total 15 hypoxic trials. Color bar, arbitrary units. (C) Correlations between voxel-wise signal changes calculated from trial 1 and trials 5, 10, and 15. All brain voxels are plotted, except low-SI regions (N indicates number of voxels). (D) Pearson’s (r) and ICC heatmap of all different paired trials for the average values of all 10 animals. (E) A schematic used for calculating contrast-to-noise ratio (CNR). All 15 repeated trials in each animal were averaged for CNR calculation. (F) CNR maps of six exemplar slices in a single animal (animal 1). (G) Voxel-wise CNR values of the cortical gray matter (GM) and the white matter (WM) corpus callosum (CC) of 3 representatives of 10 animals. The red violin plot displays the voxel-wise CNR of the cortical region [refer to the red ROI in the insert image of (E)] with 16,071 voxels, while the gray violin plot represents the CNR of the CC region with 1021 voxels underneath the selected cortical region in each animal [see green ROI in the inset mage of (E)]. (H) Animal-wise mean CNR of GM and WM averaged across all 10 animals. Paired t test was performed, ***P < 0.0001.

The repeatability of hypoxia-induced signal changes was examined across 15 trials in each animal. The voxel-wise absolute signal change induced by the hypoxic stimulation (∆S) was obtained from an average of 3-s data around the peak for each trial. ∆S maps of several representative trials are shown for a single selected slice (Fig. 3B in one mouse and fig. S2 for all nine remaining animals). Because of the use of a surface coil, low ∆S contrast was observed in ventral brain regions [see dark blue voxels in ∆S maps in Fig. 3B and fig. S3 for the low-signal regions in echo-planar images (EPI)]. As a result, a threshold was implemented with ∆S > 1 SD of baseline intensities. The ∆S values of the first trial were highly correlated with those of three randomly selected trials 5, 10, and 15 with Pearson’s r correlation coefficients (r) of 0.859, 0.848, and 0.837 and with intraclass correlation coefficients (ICC) of 0.950, 0.905, and 0.900, respectively (Fig. 3C). Excellent agreements between different pairs of trials were also observed in Bland-Altman plots (fig. S4). All r and ICC values of all paired trials were computed and presented as color heatmaps (Fig. 3D). Overall, strong correlations (r values of >0.80) and robust similarity (ICC values of >0.88) were consistently found, suggesting that the hypoxia-induced ∆S was highly reliable across repeated trials.

Since the BOLD response to hypoxic stimulus (∆S) is highly repeatable, it is expected to have a high contrast-to-noise ratio (CNR), which was examined from an average of 15 repeated trials within each animal (as illustrated in Fig. 3E). CNR maps of six exemplary slices (Fig. 3F in one representative animal and fig. S5 for the remaining nine animals) clearly differentiate the gray matter (GM) and white matter (WM) as well as the area containing large vessels. The distributions of voxel-wise CNRs in the WM corpus callosum (CC) (1021 voxels, gray violins) and in the dorsal cortical region of interest (ROI; 16071 voxels, red violins) were plotted across 3 representatives of the 10 mice (Fig. 3G). Since noise (N) was similar (0.68 ± 0.06 for GM versus 0.72 ± 0.05 for WM), CNR simply reflects baseline perfusion level. The mean voxel-wise CNR of the dorsal cortex and CC region was 7.85 ± 1.15 and 3.89 ± 0.73, respectively, with a ratio of mean CNR values between gray and WM was 2.03 ± 0.15 over 10 mice (Fig. 3H). Overall, the brief hypoxic challenge induces highly sensitive and repeatable negative BOLD responses across trials in a single session.

Repeatability of BOLD-DSC measurements across multiple imaging sessions

Since BOLD-DSC MRI is highly sensitive and repeatable in a single imaging session, we expected this high repeatability to extend across multiple sessions. To examine this, we conducted BOLD-DSC MRI measurements in 10 mice over four weekly sessions spanning a month (Fig. 4A). Throughout this period, we consistently observed almost identical hypoxia-induced BOLD responses in the primary somatosensory area from week 1 to week 4 in all animals, as depicted in Fig. 4B, which shows data from four representative animals. Notably, there were slight variations in the peak amplitudes of BOLD responses in the somatosensory area among the 10 animals over the course of four weeks (Fig. 4C).

Fig. 4. — (A) Experimental design and ROIs: BOLD-DSC measurements were conducted in 10 mice over four weekly sessions spanning a month. The hypoxic stimulus was administered under 1.5% ISO, and each subject underwent a total of 3 GE-EPI block-design runs. This identical measurement procedure was repeated every week. (B) Repeatable BOLD time courses acquired in the bilateral primary somatosensory area of four randomly chosen animals from week 1 to week 4. (C) Variations in peak BOLD signal changes in the bilateral primary somatosensory area over four weeks in 10 animals. In each animal, peak signal changes for the last 3 weeks were normalized to the peak intensity (PI) of the first week. (D and E) Quantitative CBV and CBF values measured in the WM CC, GM bilateral primary SS, and TH [see ROIs in (A)] over 4 weeks. Data were presented as mean ± SD (n = 10). One-way repeated measures ANOVA with Bonferroni correction analysis tests were performed.

Hypoxia-induced BOLD responses were converted into absolute CBV and CBF values with animal-specific corrected arterial input function (AIF) time courses (see fig. S6, A to G). Partial volume effect of an AIF was corrected by the venous output function (VOF) (26, 27). Regional CBF and CBV were computed for the CC, the primary SS, and thalamus (TH) in each animal (see ROIs in Fig. 4A), and ICC was calculated between quantitative perfusion values at the first week and following 3 weeks. The significantly high ICC values were reported for CBV (0.819, 0.822, and 0.893 in CC; 0.874, 0.961, and 0.928 in SS; and 0.922, 0.847, and 0.914 in TH for the second, third, and fourth week, respectively) and CBF (0.853, 0.914, and 0.966 in CC; 0.798, 0.792, and 0.948 in SS; and 0.857, 0.874, and 0.913 in TH for the second, third, and fourth week, respectively), indicating a high reliability in quantitative regional CBV and CBF values across four weekly sessions (Fig. 4, D and E). In summary, our BOLD-DSC measurements have demonstrated high repeatability and consistency over multiple scan sessions.

Comparison between CBV measured by endogenous BOLD-DSC and relative CBV with exogenous contrast agents

Our next question was to determine whether the CBV values determined by BOLD-DSC MRI are consistent with relative CBV measured using monocrystalline iron oxide nanoparticles (MION) as a criterion standard (28–30). We conducted BOLD-DSC MRI before and after intravenous injection of 5 mg Fe/kg in six mice under 1.5% ISO anesthesia, computing BOLD-DSC CBV and CBV-related MION-induced $∆ R_{2}^{*}$ ( $∆ R_{2, MION}^{*}$ ) (see Materials and Methods). Group-averaged maps of CBV and $∆ R_{2, MION}^{*}$ showed remarkable similarity (Fig. 5A), and voxel-wise Pearson’s r correlation coefficients in the primary SS, TH, CC, and the whole brain were high (Fig. 5B). In addition, high voxel-wise correlation values were also found in individual animals (Fig. 5C). In general, the BOLD-DSC method with 5-s transient hypoxia yields relative CBV values with high confidence.

Fig. 5. — (A) Group-averaged maps based on BOLD DSC and relative CBV maps derived from steady-state contrast enhanced MRI ( $∆ R_{2, MION}^{*}$ ) (n = 6). Ten representatives over 20 slices are presented in an interleaved manner. (B) Voxel-wise scatterplots of absolute CBV and relative CBV in SS, TH, and CC (see ROIs defined in Fig. 4A), as well as the whole brain (WB). Each red dot represents a voxel (N indicates the number of voxels), and a red line indicates the best-fitted linear line. The Pearson’s correlation (r) coefficient was calculated to illustrate the correlation of the two group-averaged measurements. (C) Voxel-wise Pearson’s r values for all six individual animals in the four ROIs. Data were presented as mean ± SD (n = 6).

Mapping CBF and CBV with BOLD-DSC MRI under different anesthetic conditions

Hypoxia-induced BOLD MRI signal changes were measured under various anesthetics (DEX + ISO, 1% ISO, 1.5% ISO, and 2% ISO) and subsequently converted into absolute CBV and CBF values with animal-specific corrected AIF (see Materials and Methods and fig. S6). AIFs were consistent for these five different anesthesia conditions (fig. S6, H and I), indicating a similar arterial dHb input under different anesthetic conditions.

Total CBV and CBF measured by GE-BOLD data collection contain large vessels (red voxels indicated by red arrows), which are mostly removed using SE-BOLD data detection (31–33). Thus, microvascular CBV and CBF measured by SE MRI are better indices of tissue perfusion (Fig. 6A). Ventral brain regions (bluish color) as well as the olfactory bulb, hypothalamus, and cerebellum areas have high quantification errors due to its low coil sensitivity (see fig. S3). To observe general patterns of CBV and CBF at four anesthetic conditions, selected slices of group-averaged total- and micro-CBV and CBF maps were presented for better visualization (red rectangle box in Fig. 6, A and D to G).

Fig. 6. — (A) Group-averaged whole-brain total- and micro-vascular CBV and CBF maps obtained from GE and SE BOLD MRI under 1.5% ISO anesthesia, respectively. Our BOLD-DSC data consisted of 20 continuous slices ranging from bregma +3.83 to −5.67 mm. To better observe general patterns of CBV and CBF at four anesthetic conditions, three slices [bregma −3.17 mm, −1.77 mm, and + 0.83 mm, marked with a red rectangle in (A)] of perfusion maps were selected for further display. Red arrows indicate the large vessels. Refer to figs. S7 and S8 for the whole-brain perfusion maps computed under other anesthetic conditions. (B and C) Corresponding atlas and anatomical T₂ images of the three selected slices. A one-hemisphere brain atlas is color-coded for ROIs. (D and E) Representative slice maps of group-averaged total-vascular CBV and CBF and bar plots of whole-brain total CBV and CBF values under four anesthetics (DEX + ISO, n = 10; 1% ISO, n = 9; 1.5% ISO, n = 10; 2% ISO, n = 10). (F and G) Representative slice maps of micro-vascular CBV and CBF and bar plots of whole-brain micro CBV and CBF values under four anesthetics (DEX + ISO, n = 7; 1% ISO, n = 7; 1.5% ISO, n = 7; 2% ISO, n = 7). Data were presented as mean ± SD. One-way repeated measures ANOVA with Bonferroni correction analysis tests were performed, *P < 0.05 and **P < 0.001.

The total and micro CBV values in the whole brain measured under DEX + ISO (3.32 ± 0.36 and 2.16 ± 0.10 ml/100 g, respectively) were slightly lower than those under 1% ISO (3.61 ± 0.60 and 2.24 ± 0.18 ml/100 g, respectively) and significantly lower than those measured under 1.5% ISO (3.95 ± 0.40 and 2.58 ± 0.27 ml/100 g, respectively) and 2% ISO (4.29 ± 0.27 2.60 ± 0.32 ml/100 g, respectively) (Fig. 6, D and F). The total and microvascular CBF values in the whole-brain under DEX + ISO (94.2 ± 8.1 and 74.4 ± 4.0 ml/100 g per min, respectively) were slightly lower than those under 1% ISO (110.7 ± 18.5 and 71.4 ± 7.7 ml/100 g per min, respectively) and significantly lower than those measured under 1.5% ISO (145.1 ± 16.8 ml/100 g per min and 87.4 ± 8.4 ml/100 g per min, respectively) and 2% ISO (161.4 ± 15.1 and 97.1 ± 12.7 ml/100 g per min, respectively) (Fig. 6, E and G). Overall, perfusion increases with ISO dose.

Region-specific effects of anesthesia on perfusion metrics assessed by BOLD-DSC MRI

To characterize regional CBV and CBF values across different anesthesia protocols, 30 bilateral brain regions were chosen on the basis of the Allen Mouse Brain Atlas (Fig. 7A for 30 ROIs) (34). Regional CBV and CBF values previously measured by the similar experimental protocol under ketamine/xylazine (KET + XYL) (4) were also included for comparisons. Heat plots of perfusion values are shown in Fig. 7B, and quantitative values are tabulated in tables S2 and S3. In general, regional CBV and CBF were lowest under DEX + ISO and highest under 2% ISO. CBV and CBF values under KET + XYL anesthesia were between those measured under DEX + ISO and 1.5% ISO, and CBV and CBF values under ISO increased with the dose of ISO.

A fraction of microvascular volume (flow) can be estimated by a ratio of micro-CBV (micro-CBF) to total-CBV (total-CBF) (Fig. 7B, bottom maps); macrovessel-dominant regions (less than 0.5, black rectangle boxes in Fig. 7B, bottom maps, blue) are anterior cingulate area (ACA), infralimbic area (ILA), dentate gyrus (DG), and pallidum (PAL), while microvessel-dominant regions (greater than 0.75, violet rectangle boxes) are hippocampal area CA and WM CC. The azygos pericallosal arteries contribute to ACA and ILA values (35), longitudinal and transverse hippocampal vessels (including arteries and veins) contribute to DG value (35, 36), and the lateral striate arteries contribute to PAL value (35). However, regional total versus microvascular CBV/CBF values are highly correlated regardless of the anesthetics used (Fig. 7C). Similarly, regional CBF and CBV are highly correlated (Fig. 7D). When a power function was used for fitting, total-CBV = 0.10 × total-CBF^0.74 (r = 0.936) and micro-CBV = 0.11 × total-CBF^0.69 (r = 0.885). A linear function was also well-described as total-CBV = 0.029 × total-CBF (r = 0.936) and micro-CBV = 0.029 × micro-CBF (r = 0.885).

Anesthetic dependence on regional CBF/CBV values was further investigated in specific ROIs (Fig. 7, E to H). Several interesting observations can be found. (i) A total of 1.5% ISO mostly induced higher CBV and CBF than KET + XYL and DEX + ISO, except the DG region. The hippocampal DG area has lower CBV/CBF under ISO compared to KET + XYL anesthesia. (ii) Despite similar CBV under DEX + ISO and KET + XYL, lower CBF values were observed under DEX + ISO. (iii) CBV and CBF increased globally with increased ISO dose but larger magnitude in CBF. (iv) Mice anesthetized with ISO had significantly higher total-CBV and total-CBF values in the subcortical TH compared to the cortical somatosensory areas (see tables S2 and S3), but micro-CBV and micro-CBF were similar. In general, our results indicate that regional perfusion values are smaller in DEX + ISO anesthesia, followed by KET + XYL and 1% ISO, and larger under ISO at 1.5 and 2% concentrations.

DISCUSSION

We have developed a gas delivery system for BOLD-DSC studies performed with commonly used injectable and volatile anesthetics. This setup enables comprehensive and quantitative assessments of cerebral perfusion at both the whole brain and regional levels, even under different anesthetic conditions. The BOLD signal change exhibits high sensitivity and repeatability across multiple trials, allowing for the determination of perfusion metrics without extensive averaging. The fidelity of our BOLD-DSC measurement is further validated by its consistency with the relative measurement derived by the standard MION contrast agent-based MRI. By using single-shot gradient-echo echo-planar image (GE-EPI) and spin-echo echo-planar image (SE-EPI) techniques, our data have provided the high-resolution anesthesia-dependent perfusion values of both the total and small-sized vasculature. These absolute measurements serve as excellent references for understanding and interpreting physiological and functional studies in mice conducted under various anesthetic conditions.

Precise hypoxic challenge under volatile anesthesia

In our setup, two gas delivery lines were adopted. The utilization of two separate input gas lines enables fast delivery of hypoxia without mixing between two gases, improving the precision of gas delivery and repeatability of BOLD MRI responses. To maintain a constant gas delivery rate during switching between two gases, it is crucial to minimize pressure buildup during the “off” period by using a commercial gas delivery system or continuously flowing gas into a scavenger or exhaust system. When a single gas line is used, then the hypoxic gas delivery (in our case, 100% N₂) is delayed due to running a long line to the animal inside the magnet (4). If a hypoxic gas is delivered through a vaporizer, then further delays and dispersion may be expected (37).

Switching off ISO during 5-s hypoxic stimulation does not cause any notable changes in physiology. Stimulus-driven BOLD MRI response is much greater with the use of anoxia stimulus than apnea (11.2% versus 1.5%). Although apnea-driven BOLD response might potentially be used, we did only use the highly sensitive anoxic stimulus for our BOLD-DSC MRI studies.

As an alternative to the hypoxic stimulus, CO₂ or O₂ challenge may be considered for perfusion studies. Both CO₂ and O₂ challenges induce BOLD responses mostly in the venous vasculature (38–40) but much less at the large arteries due to a high baseline oxygenation level of ~0.98. Thus, determining the arterial input function becomes problematic. In addition, hypercapnia induces a substantial increase in CBF and CBV (41–44), making it challenging to dissociate between baseline CBF/CBV and the evoked response. Consequently, hypercapnia or hyperoxia may be used for a qualitative measure of perfusion, which is heavily weighted toward the venous vasculature.

Repeatability and sensitivity of BOLD response and perfusion measurements

Our data have shown excellent repeatability and sensitivity of hypoxic-induced signal changes in BOLD data both within single session and multiple sessions. This strongly suggests that BOLD-DSC data obtained from a single hypoxic trial can indeed provide reliable information for CBV and CBF measurements. The high repeatability and sensitivity of our technique indicate its potential for routine perfusion measurements with high repeatability, which is particularly promising when considering the challenges faced by other techniques such as DSC with Gd injection and ASL (45, 46).

Highly sensitive BOLD DSC can be translated to human studies. In the human applications, a longer stimulus (25 s, ~5 breaths) was previously adopted at 3 T (3), leading to 16.5% arterial oxygen saturation (S_aO₂) drop (~82% S_aO₂) and −7.9% BOLD response with TE of 50 ms in the GM. If the same experimental condition was repeated at 7 T, the BOLD response with TE of 30 ms is expected to be −7.9% × (7 T/3 T) × (30 ms/50 ms) = −11%, which is similar to the magnitude of our 9.4 T measurements with TE of 11.6 ms for the 5-s stimulus.

Our BOLD-DSC method, using dHb as an endogenous contrast agent, is advantageous over conventional MRI and CT perfusion methods that rely on exogenous blood contrast agents/media. As mentioned already, the hypoxic gas challenge is easily achievable and repeatable without concerning recirculation and leakage of long blood half-life exogenous contrast agents (>30 min), which simplifies DSC modeling (4). Unlike Gd-DTPA, dHb is insensitive to T₁ (47), which simplifies data collection with T₂/ $T_{2}^{*}$ weighting.

Investigation of the effects of commonly used anesthetics in animal research

Different anesthetics have distinct mechanisms of action and produce different effects on both neuronal and vascular activity (13). Ketamine is a noncompetitive NMDA (N-methyl-d-aspartate receptor) antagonist that affects both excitatory and inhibitory neurons, and has a minor effect on cerebral vasodilation (48, 49). On the other hand, vasodilative ISO potentiates γ-aminobutyric acid type A receptor and inhibits NMDA receptor (50, 51). Dexmedetomidine and xylazine are vasoconstrictive α2 agonists. The effects of these anesthetics were previously investigated in resting state functional MRI [fMRI DEX + ISO (16, 52)], evoked fMRI [DEX + ISO and KET + XYL (14, 16, 53–55)], and biomedical MR research [ISO (51, 56)]. However, the effects of these anesthetics on regional and whole-brain perfusion in mice have not been extensively investigated.

Our results revealed higher CBV and CBF levels under high-dose ISO compared to low-dose. At a high dose, ISO acts as a globally robust vasodilator due to its dose-dependent effect on vasodilation (12, 57). In addition, we found similarities in baseline CBV and CBF between low-dose ISO at 1% and KET + XYL. Furthermore, our perfusion measurements under DEX + ISO were lower compared to those under KET + XYL. This can be explained by the stronger vasoconstrictive effect of DEX compared to XYL (58–60). Our total CBF values (94.2 ml/100 g per min for DEX + ISO and 110.7, 145.1, and 161.4 ml/100 g per min for 1.0, 1.5, and 2% ISO, respectively) agree well with quantitative CBF values measured with ASL MRI (19, 20, 61); Munting et al. (19) measured CBF of three 1.5-mm-thick coronal slices in 10 month-old mice and found mean CBF of 157.7 ml/100 g per min under 1.5 to 2% ISO, 138.9 ml/100 g per min under 1.25% ISO, and 84.4 ml/100 g per min under DEX. Note that Munting et al. (19) used DEX (0.30 mg/kg per hour), while we used DEX (0.05 mg/kg per ) and vasodilator 0.3% ISO. In later age-dependent CBF measurements under 1.25% ISO (20), CBF significantly decreased from 155 ml/100 g per min at 3 months old to 121 ml/100 g per min at 6 months old. Zheng et al. (61) measured CBF of one 1-mm-thick coronal slice in ISO-anesthetized mice and found an average CBF of 158 ml/100 g per min under 1.2% ISO and 178 ml/100 g per min under 1.8% ISO. An increase in CBF with an increased ISO dose was also found in laser Doppler flow and speckle measurements of mouse cortical surface (62, 63); Takuwa et al. (62, 63) found an 18% CBF increase under 1.5% ISO compared to the awake condition, and Sullender et al. (62, 63) found an 85% CBF increase at 2% ISO relative to awake. In our perfusion studies under ISO, CBF in the TH is much higher than that in the cortical area, consistent with previous observations [see figure 4 in Zheng et al. (61), figure 1 in Munting et al. (19), and figure 2 in Munting et al. (20)].

Potential limitations

Quantification of perfusion involves multiple sources of errors, similar to Gd-DSC MRI (64–66), except T₁ effects, leakage, and recirculation. Underestimation of the AIF (leading to overestimation of perfusion values) can happen if the arterial/venous response reaches to the noise level or partial volume effects are not corrected properly. The AIF is additionally sensitive to the orientation of the artery. In our experimental condition with short TE, the peak intensity was much higher than the noise level, thus the saturation of AIF and VOF was not existent. We also assumed that the venous voxel contained only the vascular component without the contamination of nearby tissue, which was confirmed in our previous studies obtained with the same imaging parameters (4). Consequently, the corrected AIF was reasonable.

In the BOLD-DSC modeling, we assumed a linear relationship between dHb concentration and Δ $R_{2}^{*}$ . The conversion constant from dHb concentration to Δ $R_{2}^{*}$ was assumed to be identical for both tissue and blood when quantifying perfusion using BOLD-DSC MRI. However, it is well-known (33, 67) that BOLD MRI contrast has a complex biophysical mechanism originating from both intravascular and extravascular components. Intravascular Δ $R_{2}^{*}$ is quadratically dependent on both magnetic field and oxygenation change (ΔY), while extravascular Δ $R_{2}^{*}$ is linearly dependent on magnetic field, ΔY and CBV (33, 67). For voxel sizes much smaller than large artery and venous diameters, arterial and venous blood signals are predominantly of an intravascular origin, while tissue signals contain both intravascular and extravascular components at clinical fields of 1.5 to 3 T but are dominated by the extravascular component at ultrahigh fields of >7 T [figure 5 in Uludağ et al. (33)]. Thus, the conversion constant is dependent on magnetic field strength and the extravascular/intravascular composition.

Recently, Schulman et al. (68) investigated the accuracy of DSC quantification by using computer simulations and human studies at 3 T. Δ $R_{2}^{*}$ in the artery (8 to 10 s⁻¹) has been shown to be much lower than that in the vein (30 to 40 s⁻¹) for BOLD-DSC MRI at 3 T [figure 3 in Vu et al. (3) and figure 3 in Schulman et al. (68)], which is attributed to the nonlinear dependency of Δ $R_{2}^{*}$ on the vessel’s baseline oxygenation, despite the expected consistency in ΔY across cerebral tissue and blood. In our data (see Fig. 2D), Δ $R_{2}^{*}$ in the artery (~150 s⁻¹) is slightly higher than that in the vein (~120 s⁻¹), indicating an approximately linear dependency of Δ $R_{2}^{*}$ on the oxygenation level at 9.4 T. Please note that for small voxels (i.e., the absence of large- and medium-scale susceptibility gradient within the voxel), intravascular $R_{2}^{*}$ and R₂ as a function of blood oxygenation should be almost identical. On the basis of the relationship between Y and blood R₂ at 9.4 T [R₂ (s⁻¹) = 478 to 458Y (69)], blood R₂ is expected to be ~30 s⁻¹ and ~200 s⁻¹ for Y of 0.98 and 0.6, respectively, which agree well with our arterial blood $R_{2}^{*}$ data (baseline arterial blood $R_{2}^{*}$ of ~40 s⁻¹ and ${Δ R}_{2}^{*}$ of ~150 s⁻¹ in Fig. 2). To quantify CBF and CBV accurately, it is crucial to determine the conversion constants from ΔY to Δ $R_{2}^{*}$ in blood as a function of baseline Y and to Δ $R_{2}^{*}$ in tissue with various compositions of different vascular diameters and densities. To that end, systematic simulations and field-dependent BOLD-DSC experiments are needed in the future.

Additional potential limitations in our BOLD-DSC MRI studies include several factors. First, the use of a 10–mm–inner diameter RF surface coil reduces sensitivity in deeper ventral regions. This limitation could be addressed by using a high-sensitivity array or a homogeneous coil (70). Second, a temporal resolution of 1 s was used for whole-brain mapping. Increasing the temporal resolution can accurately depict BOLD response curves, thereby improving the accuracy of CBF quantification (23). To achieve higher temporal resolution, simultaneous multislice sampling or compressed sensing MRI can be applied (71–73). Third, hypoxia can induce physiological changes. Breathing 100% N₂ for 5 s leads to the dilation of cortical vessels up to 11% (4), resulting in the overestimation of baseline CBV. To mitigate hypoxia-induced physiological changes, a milder hypoxic stimulus is considered.

As described, the accuracy of perfusion metrics is closely related to various experimental parameters and BOLD modeling, similar to Gd-DSC measurements (64–66). Although quantification of CBV and CBF is susceptible to the aforementioned issues, relative perfusion values remain valid.

In conclusion, the BOLD-DSC MRI approach with transient hypoxic stimulus is easily implementable, quantitative, highly sensitive, and repeatable, allowing biomedical scientists to longitudinally and repeatedly map cerebral perfusion in normal, diseased, and transgenic mice. This approach will be viable to identify early biomarkers, monitor disease progression and evaluate treatment efficacies in the same animal and to assess vascular physiology in numerous transgenic mouse models for linking between genes and physiological changes. A similar approach is feasible in humans (3, 5, 68), facilitating the translation of findings from animal research to human studies.

MATERIALS AND METHODS

Animal preparations

All animal experiments were approved by the Institutional Animal Care and Use Committee of Sungkyunkwan University. The experiments adhered to the standards for humane animal care and use as set by the Animal Welfare Act in compliance with the Animal Research: Reporting of in Vivo Experiments (ARRIVE) guidelines. Sixty-five male mice (C57BL/6J, weighing 25 to 28 g, and aged 2 to 5 months old, Orient Bio, Korea) were used. Detailed information regarding to animal uses can be found in table S1. Exclusions were made for three mice in the DEX + ISO group (which were not stable due to no proper intravenous infusion of DEX) and one mouse in the 1% ISO group (which was awake).

MRI acquisitions

All MRI experiments were performed on a Bruker Biospec 9.4 T/30-cm horizontal bore instrument with an actively shielded 12.0-cm-diameter insert operating at a maximum gradient strength of 66 Gauss/cm and a rise time of 141 μs. The mouse brain was positioned as close as possible to the isocenter of the magnet. A quadrature birdcage coil (86 mm inner diameter) was used for excitation, and a 10-mm receiver surface coil was positioned on top of the mouse head. The magnetic field homogeneity was globally shimmed and then a local shim was optimized using the MAPSHIM protocol with an ellipsoid shim volume covering the cerebrum (ParaVision 6.0.1, Bruker BioSpin, USA).

In general, single-shot GE-EPI were acquired for total-vasculature perfusion mapping with the following parameters: repetition time (TR)/TE = 1000/11.6 ms (except TE-dependent studies), flip angle = 50°, receiver bandwidth = 300 kHz, spatial resolution = 156 μm × 156 μm × 500 μm, 20 continuous slices with an interleaved order, slice thickness = 500 μm, and dummy scans of 10. Alternatively, single-shot SE-EPI were used for microvascular perfusion mapping, using the same parameters except for TE = 22 ms and flip angles = 90°/180°.

During the experiment, physiological data were monitored using a physiological monitoring system (Model 1030, Small Animal Instruments Inc., USA) and recorded using a data acquisition system (Biopac Systems Inc., USA). The heart rate and peripheral oxygen saturation values were monitored using an MR-compatible optical pulse oximeter attached to the tail. A respiration pad was used to measure respiration rates. End-tidal CO₂ was monitored using a multiparameter physiological monitor (LifeWindow LW9x, Digicare Biomed Tech, USA). The temperature of the mice was maintained at 37° ± 0.5°C using a circulating warm water blanket.

Anesthetic regimes

Four anesthesia conditions were examined in this study.

Isoflurane with three different doses (1, 1.5, and 2.0% ISO)

Mice were initially anesthetized with 3.5 to 4% ISO in a mixture of oxygen:air of 2:3 for 3 to 4 min. The ISO concentration was then lowered to 2% for the MRI setup. Subsequently, the ISO level was adjusted and maintained at a target level of 1, 1.5, and 2.0%. Note that one minimum alveolar concentration of ISO in mouse is approximately 1.34% (74). Data acquisitions commenced 15 to 20 min after adjusting the ISO dose to allow sufficient time for the new state of anesthesia to be stabilized (51, 75, 76). The order of varying ISO dose was randomly assigned, and all experiments were complete within 4 hours from the first acquisition in each mouse.

Dexmedetomidine (with 0.3% ISO supplement) protocol (DEX ± ISO)

With an ISO concentration of 2%, a 31-gauge needle was cannulated in the tail vein for intravenous administration of anesthesia using a syringe pump (Harvard Apparatus Standard Infuse/Withdraw PHD ULTRA). After transferring the mouse into the magnet, an intravenous bolus of DEX (0.05 mg/kg) was administered, and ISO was halted. An intravenous infusion of 0.05 mg/kg per hour was initiated after 10 min. ISO administration was reinstated at a concentration of 0.3% for 30-min period following the intravenous bolus (60).

Setup and stimulus designs

Two inhaled gas mixtures were used, one for control medical gas with anesthetics and the other for the hypoxic stimulus. These gases were connected to a breathing cone, and a transistor-transistor logic signal synchronized with the MRI scanner was used to switch between the two gases accurately (see Fig. 1A). To maintain a consistent gas pressure level during experiments, a solenoid two-way pinch valve was used to control the flow of medical gas to the animal, while a three-channel programmable gas mixer (GSM-3 Gas Mixer, CWE Inc. Ardmore, PA 19003) was used to control the hypoxic gas stimulus. Exhaled and residual gas within the nose cone was efficiently removed by maintaining a slightly negative pressure in the exhaust gas line.

The normoxic baseline was set to 40% O₂ with balanced N₂ to ensure adequate oxygenation during anesthesia, while the transient hypoxic stimulus, conducted without any inhaled anesthetics, used 100% N₂. Each run, lasting 5 min and 45 s, consisted of 20-s normoxia followed by a cycle of (5-s hypoxia to 60-s normoxia) repeated five times (as shown in Fig. 1B). For each anesthetic condition in each animal, three runs (15 trials of 5-s hypoxia) were performed to gather data for total vasculature-sensitive GE-EPI, and five to six runs (25 to 30 trials) were conducted to collect data for microvascular-sensitive SE-EPI. Seven main experiments were performed in this study (see table S1).

Experiment #1: Potential physiological changes caused by the absence of ISO during hypoxic stimulation

During the 5-s hypoxic stimulus period, no inhaled anesthetic was administered, which might modulate physiology. To assess the impact of the absence of ISO during the stimulation period, a 5-s normoxic stimulus (40% O₂ and 60% N₂, the same as the baseline condition) was used in place of 100% N₂ gas. Five mice under 1.5% ISO were subjected to both normoxic and hypoxic stimuli (Fig. 1, C and D). A total of three GE-EPI block-design runs were collected in each animal.

Experiment #2: Sensitivity of BOLD responses induced by varying durations of apnea stimulus

To assess the viability of apnea for perfusion studies, we examined the sensitivity induced by three different apnea periods: 5, 10, and 15 s. Initially, mice self-breathed the delivered baseline gas mixture (40% O₂ and 60% N₂). Then, the apnea was initiated by discontinuing the delivery of the baseline gas. Five mice under 1.5% ISO were subjected to the apneic stimulus. A total of three GE-EPI block-design runs were conducted for each animal.

Experiment #3: TE-dependent hypoxia-induced transverse relaxation rate change $∆ R_{2}^{*}$

To characterize the TE dependency of the arterial and tissue concentration time curves, we used single-shot GE-EPI with varying TEs (TR/TE = 1000/11.6, 13, 15, 17, and 20 ms). Hypoxic stimulus was administered under 1.5% ISO in six animals. Each animal underwent a total of three GE-EPI block-design runs.

Experiment #4: Repeatability of weekly BOLD-DSC measurements

To assess the reliability of perfusion metric quantification across sessions, we conducted BOLD-DSC measurements with 10 subjects over four weekly sessions spanning a month. The identical procedure was repeated every week. Hypoxic stimulus was administered under 1.5% ISO. Each subject underwent a total of three GE-EPI block-design runs.

Experiment #5: Comparison between CBV measured by BOLD DSC and relative CBV with MION contrast agents

To compare baseline CBV values determined by BOLD-DSC and relative CBV measured using MION, three runs of GE-EPI BOLD MRI were conducted on six mice before and after intravenous injection of MION (5 mg Fe/kg) (77) (Feraheme, AMAG Pharmaceuticals, USA). Experiments were carried out under 1.5% ISO.

Experiment #6: BOLD-DSC measurements under different doses of ISO

Twenty mice were included in the BOLD-DSC studies, which were performed under three distinct anesthetic protocols: 1.5% ISO (n = 10), 1.0% ISO, and 2.0% ISO (n = 10). GE-EPI DSC-MRI perfusion data were collected in all mice, while SE-EPI data collection was used in most animals (see table S1). However, one mouse under 1% ISO was awake during experiments and thus was not included.

Experiment #7: BOLD-DSC measurements under dexmedetomidine

Thirteen mice were enrolled in the BOLD-DSC studies, which were conducted under DEX + ISO anesthesia. The DSC-MRI perfusion data were acquired using both the GE-EPI and SE-EPI sequences. Note that three mice were excluded from the analysis because they remained awake and unstable due to improper DEX infusion.

Data processing

All BOLD time series data for each mouse were analyzed using MATLAB (MathWorks, USA), the Analysis of Functional Neuroimages package (78). For individual EPI images, the following preprocessing steps were performed: slice timing correction, motion correction, and linear detrending for signal drift removal. The group-averaged CBV and CBF maps were generated on the mouse brain template space through multiple steps. First, individual EPI images (as shown in fig. S3, 156 μm × 156 μm × 500 μm) were coregistered to the individual anatomical T₂ images (78 μm × 78 μm × 500 μm) using a linear transformation. Second, the T₂-weighted images of all individual subjects were normalized and averaged while applying linear and nonlinear transformations to generate a mouse brain template. Third, all EPI images coregistered in the first step were normalized to the mouse brain template created in the second step (78 μm × 78 μm × 500 μm) using the linear and nonlinear transformation. Fourth, the individual CBV and CBF maps were coregistered to the individual T₂ images using the transformation parameter obtained in the first step and then normalized onto the mouse brain template using the transformation parameters obtained in the second step. Last, the Allen Mouse Brain Atlas was also registered to the brain template using both nonlinear and linear transformations.

Quantification of perfusion metrics from arterial and tissue response curves

Since the hypoxic stimulus induces large and repeatable GE-BOLD responses, CBV, CBF, and mean transit time can be quantified with the DSC tracer kinetic theory (4, 79–81). The signal loss in tissue and arteries by a short hypoxic challenge was captured by using dynamic $T_{2}^{*}$ -weighted GE-EPI (or T₂-weighted SE-EPI) (fig. S6, A and D) (4). Then, signal intensity (SI) change of each voxel was converted to the relaxation rate change $∆ R_{2}^{*}$ ( $∆ R_{2}$ ) by $∆ R_{2}^{*}$ = −hypoxia-induced relative change/TE (fig. S6, B and E), since it is linearly related with dHb concentration [C_t(t) as: C_t(t) = k $\cdot ∆ R_{2}^{*}$ where k is constant] (81, 82).

With the arterial as well as the total tissue concentration as a function of time, absolute CBV can be calculated as the ratio of the area under the tissue (AUC1, fig. S6C) and corrected AIF time curves (AUC2, fig. S6F). Note that the partial volume effect of an arterial voxel response was corrected by the normalization with a VOF for obtaining the corrected AIF (26, 27). The tissue concentration response is proportional to the amount of blood (delivering arterial tracer concentration, AIF) passing through the tissue per unit time (CBF) and can be defined as the convolution of the tissue response function [CBF × residue function R(t) and AIF as C_t(t) = CBF · AIF⨂R(t)] (83, 84). The tissue impulse response function was obtained by deconvolution using the traditional singular value decomposition approach with a fixed threshold (20% cutoff) (85–87). At R(t) = 0, CBF was determined as the initial height of the tissue impulse response function (fig. S6G).

To assess the regional CBV and CBF, we defined 30 bilateral brain regions based on the Allen Mouse Brain Atlas (34). The brain regions included CC, midbrain; TH, hypothalamus, pons, medulla, cerebellar cortex, cerebellar nuclei, striatum, PAL,olfactory areas, visceral area, visual areas, temporal association areas, somatosensory areas, retrosplenial area, posterior parietal association area, prelimbic area, perirhinal area, orbital area, somatomotor areas, ILA, gustatory areas, ectorhinal area, auditory areas, agranular insular area, ACA, DG, and cornu ammonis (Fig. 7A).

Suitability of the setup for BOLD-DSC with ISO (experiments #1 and #2)

To address the concern regarding the potential physiological changes caused by the absence of ISO during the 5-s hypoxia period, we calculated and compared the BOLD percent change obtained from primary somatosensory area using GE-EPI under both hypoxic and normoxic conditions (GE-hypoxia and GE-normoxia, respectively).

Furthermore, to assess the feasibility of using apneic stimulus for BOLD DSC, we also computed the BOLD percent change during 5-s apnea periods (GE-apnea). The mean percent change was derived from all hypoxic, normoxic, and apneic episodes. Subsequently, one-way measures analysis of variance (ANOVA) with Bonferroni-corrected post hoc analysis tests were performed to compare the BOLD percent changes between GE-hypoxia, GE-normoxia, and GE-apnea.

TE dependence of the hypoxia-induced transverse relaxation rate changes (experiment #3)

We used an automatic algorithm for selecting candidate arterial and venous voxel, while tissue voxel was chosen from the somatosensory region. To quantify background noise levels, we determined the SD of the baseline signal intensities over a 20-s period in each individual hypoxic trial (with a total of 15 trials) and then calculated the average for each TE condition. The peak was identified as the maximum signal magnitude in each individual trial. Baseline SNR was calculated from 20-s prestimulus signal intensities as SI/noise, where SI is the average baseline signal intensities and noise is the average SD of the signal intensities. The baseline $T_{2}^{*}$ values for arterial and venous blood and brain tissue were determined by fitting the GE-BOLD signal intensities acquired at five TE values during the 20-s prestimulus period to an exponential decay function. To determine hypoxia-induced $∆ R_{2}^{*}$ at different TEs, we used a linear relationship between fractional signal change (∆S/S₀) and TE: $∆ S / S_{0} = - TE \times ∆ R_{2}^{*}$ , where ∆S is the hypoxic-induced signal change, S₀ is the mean baseline SI, and TE is the TE. Subsequently, one-way repeated measures ANOVA with Bonferroni-corrected post hoc analysis tests were performed.

Repeatability and detectability of hypoxic response across the whole brain within a single imaging session (experiment #5)

To determine the repeatability and detectability of responses, data acquired under 1.5% ISO were used for the following analyses:

1) To assess the reliability of hypoxia-induced BOLD signal change (∆S, averaged over three volumes around the peak) across trials and runs, we computed the Pearson’s correlation (r) and ICC (95% confidence interval, P < 0.001) between voxel-wise ΔS values of two different trials. The agreement between two different variables was evaluated using Bland-Altman plots. All r and ICC values of all possible paired trials were computed within each individual animal.

2) Voxel-wise CNR, as an index of detectability, was computed by dividing the mean of SI change ∆S by the SD of the baseline signal intensities (defined as noise, N) over a 20-s period for each animal (Fig. 3E). All 15 trials in each animal were first averaged, and voxel-wise CNR was calculated from the averaged time series data. To determine the CNR of GM and WM, along with their ratio, across 10 mice, we computed the mean CNRs of all voxels in the cortical area and the CC ROI (highlighted in red and green, respectively in Fig. 3E insert image). Subsequently, a paired t test was performed to determine the statistical significance of differences between these two regional CNR values in each animal.

Repeatability of BOLD-DSC measurements across multiple imaging sessions (experiment #4)

To comprehensively assess the repeatability of BOLD-DSC measurements across multiple sessions, we conducted the following analyses:

1) Repeatability of hypoxia-induced BOLD responses in the primary SS over four consecutive weeks within each subject. Peak amplitudes were calculated as the average of three data points around the peak. Subsequently, we normalized the peak amplitudes of BOLD responses from week 2 to week 4 by those of week 1.

2) Repeatability of regional CBV and CBF measurements within targeted ROIs (CC, primary somatosensory, and TH) across sessions. We determined total-vasculature perfusion metrics in 10 animals over four consecutive weeks. ICC values were calculated from two different pairs: week 1 versus week 2, week 1 versus week 3, and week 1 versus week 4. Subsequently, one-way repeated measures ANOVA with Bonferroni-corrected post hoc analysis tests were performed.

Comparison between CBV measured by endogenous BOLD DSC and relative CBV with MION contrast agents (experiment #5)

Baseline CBV values were determined from GE BOLD-DSC data before MION injection (as shown in fig. S6). It should be noted that CBV after MION injection was not computed due to an inability to obtain the accurate AIF. The change in transverse relaxation rates by MION at baseline ( $R_{2, MION}^{*}$ ) corresponds to relative CBV index. Steady-state ∆ $R_{2, MION}^{*}$ was determined from 20-s prestimulus baseline signals before (S_pre) and after MION injection (S_post) as ln(S_pre/S_post)/TE.

Quantitative total and micro-vasculature perfusion metrics measured under different anesthetics (experiments #6 and #7)

Hypoxia-induced GE-BOLD signal changes were measured under various anesthetics (DEX + ISO, 1% ISO, 1.5% ISO, and 2% ISO) and subsequently converted into absolute total-vasculature CBV and CBF values with animal-specific AIF time courses (black time courses in fig. S6H). In addition, SE-BOLD data obtained during hypoxic trials were used for the quantification of micro-vasculature CBV and CBF. To calculate perfusion values with SE-BOLD detection, the average corrected AIFs estimated in GE-BOLD data for each anesthetic condition were used (red thick time courses in fig. S6H) [see also in (4)].

Acknowledgments

We acknowledge K. Uludağ and J. Schulman at University of Toronto, Toronto, Canada for providing critical inputs of BOLD-DSC modeling.

Funding: This work was supported by the Institute of Basic Science (IBS-R015-D1 to S.-G.K.).

Author contributions: Conceptualization: T.T.L. and S.-G.K. Methodology: T.T.L., S.-G.K., G.H.I., and C.H.L. Investigation: T.T.L., G.H.I., and C.H.L. Software: S.H.C. Formal analysis: T.T.L. and S.H.C. Visualization: T.T.L. Project administration: S.-G.K.. Supervision: S.-G.K.. Writing—original draft: T.T.L. Writing—reviewing and editing: S.-G.K..

Competing interests: T.T.L., G.H.I., and S.-G.K. are listed as inventors on Korean patent “Perfusion MRI method and system” (no. 10-2022-0181724, registered on 22 December 2022). C.H.L. and S.H.C. declare that they have no competing interests.

Data and materials availability: All data needed to evaluate the conclusions in the paper are present in the paper and/or the Supplementary Materials.

Supplementary Materials

This PDF file includes:

Figs. S1 to S8

Tables S1 to S3

sciadv.adm7605_sm.pdf^{(2.2MB, pdf)}

REFERENCES AND NOTES

1.L. Edvinsson, E. T. MacKenzie, J. McCulloch, Cerebral Blood Flow and Metabolism (Raven Press, 1993), vol. 1185. [Google Scholar]
2.Wintermark M., Sesay M., Barbier E., Borbély K., Dillon W. P., Eastwood J. D., Glenn T. C., Grandin C. B., Pedraza S., Soustiel J.-F., Nariai T., Zaharchuk G., Caillé J.-M., Dousset V., Yonas H., Comparative overview of brain perfusion imaging techniques. Stroke 36, e83–e99 (2005). [DOI] [PubMed] [Google Scholar]
3.Vu C., Chai Y., Coloigner J., Nederveen A. J., Borzage M., Bush A., Wood J. C., Quantitative perfusion mapping with induced transient hypoxia using BOLD MRI. Magn. Reson. Med. 85, 168–181 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
4.Lee D., Le T. T., Im G. H., Kim S.-G., Whole-brain perfusion mapping in mice by dynamic BOLD MRI with transient hypoxia. J. Cereb. Blood Flow Metab. 42, 2270–2286 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
5.Poublanc J., Sobczyk O., Shafi R., Sayin E. S., Schulman J., Duffin J., Uludag K., Wood J. C., Vu C., Dharmakumar R., Fisher J. A., Mikulis D. J., Perfusion MRI using endogenous deoxyhemoglobin as a contrast agent: Preliminary data. Magn. Reson. Med. 86, 3012–3021 (2021). [DOI] [PubMed] [Google Scholar]
6.Cryan J. F., Holmes A., The ascent of mouse: Advances in modelling human depression and anxiety. Nat. Rev. Drug Discov. 4, 775–790 (2005). [DOI] [PubMed] [Google Scholar]
7.Götz J., Ittner L. M., Animal models of Alzheimer's disease and frontotemporal dementia. Nat. Rev. Neurosci. 9, 532–544 (2008). [DOI] [PubMed] [Google Scholar]
8.Nestler E. J., Hyman S. E., Animal models of neuropsychiatric disorders. Nat. Neurosci. 13, 1161–1169 (2010). [DOI] [PMC free article] [PubMed] [Google Scholar]
9.Gehrmann J., Hammer P. E., Maguire C. T., Wakimoto H., Triedman J. K., Berul C. I., Phenotypic screening for heart rate variability in the mouse. Am. J. Physiol. Heart Circ. Physiol. 279, H733–H740 (2000). [DOI] [PubMed] [Google Scholar]
10.Janssen B. J., De Celle T., Debets J. J., Brouns A. E., Callahan M. F., Smith T. L., Effects of anesthetics on systemic hemodynamics in mice. Am. J. Physiol. Heart Circ. Physiol. 287, H1618–H1624 (2004). [DOI] [PubMed] [Google Scholar]
11.Szczęsny G., Veihelmann A., Massberg S., Nolte D., Messmer K., Long-term anaesthesia using inhalatory isoflurane in different strains of mice—The haemodynamic effects. Lab. Anim 38, 64–69 (2004). [DOI] [PubMed] [Google Scholar]
12.Iida H., Ohata H., Iida M., Watanabe Y., Dohi S., Isoflurane and sevoflurane induce vasodilation of cerebral vessels via ATP-sensitive K⁺channel activation. Anesthesiology 89, 954–960 (1998). [DOI] [PubMed] [Google Scholar]
13.Masamoto K., Kanno I., Anesthesia and the quantitative evaluation of neurovascular coupling. J. Cereb. Blood Flow Metab. 32, 1233–1247 (2012). [DOI] [PMC free article] [PubMed] [Google Scholar]
14.Weber R., Ramos-Cabrer P., Wiedermann D., Van Camp N., Hoehn M., A fully noninvasive and robust experimental protocol for longitudinal fMRI studies in the rat. Neuroimage 29, 1303–1310 (2006). [DOI] [PubMed] [Google Scholar]
15.Fukuda M., Vazquez A. L., Zong X., Kim S. G., Effects of the α2-adrenergic receptor agonist dexmedetomidine on neural, vascular and BOLD fMRI responses in the somatosensory cortex. Eur. J. Neurosci. 37, 80–95 (2013). [DOI] [PMC free article] [PubMed] [Google Scholar]
16.Grandjean J., Schroeter A., Batata I., Rudin M., Optimization of anesthesia protocol for resting-state fMRI in mice based on differential effects of anesthetics on functional connectivity patterns. Neuroimage 102, 838–847 (2014). [DOI] [PubMed] [Google Scholar]
17.Adamczak J. M., Farr T. D., Seehafer J. U., Kalthoff D., Hoehn M., High field BOLD response to forepaw stimulation in the mouse. Neuroimage 51, 704–712 (2010). [DOI] [PubMed] [Google Scholar]
18.Sinclair M. D., A review of the physiological effects of α₂-agonists related to the clinical use of medetomidine in small animal practice. Can. Vet. J. 44, 885–897 (2003). [PMC free article] [PubMed] [Google Scholar]
19.Munting L. P., Derieppe M. P., Suidgeest E., Denis de Senneville B., Wells J. A., van der Weerd L., Influence of different isoflurane anesthesia protocols on murine cerebral hemodynamics measured with pseudo-continuous arterial spin labeling. NMR Biomed. 32, e4105 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
20.Munting L. P., Derieppe M., Suidgeest E., Hirschler L., van Osch M. J., de Senneville B. D., van der Weerd L., Cerebral blood flow and cerebrovascular reactivity are preserved in a mouse model of cerebral microvascular amyloidosis. eLife 10, e61279 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
21.Muir E. R., Shen Q., Duong T. Q., Cerebral blood flow MRI in mice using the cardiac-spin-labeling technique. Magn. Reson. Med. 60, 744–748 (2008). [DOI] [PMC free article] [PubMed] [Google Scholar]
22.Smith M. R., Lu H., Frayne R., Signal-to-noise ratio effects in quantitative cerebral perfusion using dynamic susceptibility contrast agents. Magn. Reson. Med. 49, 122–128 (2003). [DOI] [PubMed] [Google Scholar]
23.Knutsson L., Ståhlberg F., Wirestam R., Aspects on the accuracy of cerebral perfusion parameters obtained by dynamic susceptibility contrast MRI: A simulation study. Magn. Reson. Imaging 22, 789–798 (2004). [DOI] [PubMed] [Google Scholar]
24.Ellinger R., Kremser C., Schocke M. F., Kolbitsch C., Griebel J., Felber S. R., Aichner F. T.et al. , J. Comput. Assist. Tomogr. 24, 942–948 (2000). [DOI] [PubMed] [Google Scholar]
25.Han S., Son J. P., Cho H., Park J. Y., Kim S. G., Gradient-echo and spin-echo blood oxygenation level–dependent functional MRI at ultrahigh fields of 9.4 and 15.2 Tesla. Magn. Reson. Med. 81, 1237–1246 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
26.Knutsson L., Börjesson S., Larsson E. M., Risberg J., Gustafson L., Passant U., Ståhlberg F., Wirestam R., Absolute quantification of cerebral blood flow in normal volunteers: Correlation between Xe-133 SPECT and dynamic susceptibility contrast MRI. J. Magn. Reson. Imaging 26, 913–920 (2007). [DOI] [PubMed] [Google Scholar]
27.Knutsson L., Lindgren E., Ahlgren A., van Osch M. J. P., Bloch K. M., Surova Y., Ståhlberg F., Van Westen D., Wirestam R., Dynamic susceptibility contrast MRI with a prebolus contrast agent administration design for improved absolute quantification of perfusion. Magn. Reson. Med. 72, 996–1006 (2014). [DOI] [PubMed] [Google Scholar]
28.Moreno H., Hua F., Brown T., Small S., Longitudinal mapping of mouse cerebral blood volume with MRI. NMR Biomed. 19, 535–543 (2006). [DOI] [PubMed] [Google Scholar]
29.Kim S. G., Harel N., Jin T., Kim T., Lee P., Zhao F., Cerebral blood volume MRI with intravascular superparamagnetic iron oxide nanoparticles. NMR Biomed. 26, 949–962 (2013). [DOI] [PMC free article] [PubMed] [Google Scholar]
30.Wu E. X., Wong K. K., Andrassy M., Tang H., High-resolution in vivo CBV mapping with MRI in wild-type mice. Magn. Reson. Med. 49, 765–770 (2003). [DOI] [PubMed] [Google Scholar]
31.Ogawa S., Menon R., Tank D. W., Kim S., Merkle H., Ellermann J., Ugurbil K., Functional brain mapping by blood oxygenation level-dependent contrast magnetic resonance imaging. A comparison of signal characteristics with a biophysical model. Biophys. J. 64, 803–812 (1993). [DOI] [PMC free article] [PubMed] [Google Scholar]
32.Boxerman J. L., Hamberg L. M., Rosen B. R., Weisskoff R. M., MR contrast due to intravascular magnetic susceptibility perturbations. Magn. Reson. Med. 34, 555–566 (1995). [DOI] [PubMed] [Google Scholar]
33.Uludağ K., Müller-Bierl B., Uğurbil K., An integrative model for neuronal activity-induced signal changes for gradient and spin echo functional imaging. Neuroimage 48, 150–165 (2009). [DOI] [PubMed] [Google Scholar]
34.Lein E. S., Hawrylycz M. J., Ao N., Ayres M., Bensinger A., Bernard A., Boe A. F., Boguski M. S., Brockway K. S., Byrnes E. J., Chen L., Chen L., Chen T.-M., Chin M. C., Chong J., Crook B. E., Czaplinska A., Dang C. N., Datta S., Dee N. R., Desaki A. L., Desta T., Diep E., Dolbeare T. A., Donelan M. J., Dong H.-W., Dougherty J. G., Duncan B. J., Ebbert A. J., Eichele G., Estin L. K., Faber C., Facer B. A., Fields R., Fischer S. R., Fliss T. P., Frensley C., Gates S. N., Glattfelder K. J., Halverson K. R., Hart M. R., Hohmann J. G., Howell M. P., Jeung D. P., Johnson R. A., Karr P. T., Kawal R., Kidney J. M., Knapik R. H., Kuan C. L., Lake J. H., Laramee A. R., Larsen K. D., Lau C., Lemon T. A., Liang A. J., Liu Y., Luong L. T., Michaels J., Morgan J. J., Morgan R. J., Mortrud M. T., Mosqueda N. F., Ng L. L., Ng R., Orta G. J., Overly C. C., Pak T. H., Parry S. E., Pathak S. D., Pearson O. C., Puchalski R. B., Riley Z. L., Rockett H. R., Rowland S. A., Royall J. J., Ruiz M. J., Sarno N. R., Schaffnit K., Shapovalova N. V., Sivisay T., Slaughterbeck C. R., Smith S. C., Smith K. A., Smith B. I., Sodt A. J., Stewart N. N., Stumpf K.-R., Sunkin S. M., Sutram M., Tam A., Teemer C. D., Thaller C., Thompson C. L., Varnam L. R., Visel A., Whitlock R. M., Wohnoutka P. E., Wolkey C. K., Wong V. Y., Wood M., Yaylaoglu M. B., Young R. C., Youngstrom B. L., Yuan X. F., Zhang B., Zwingman T. A., Jones A. R., Genome-wide atlas of gene expression in the adult mouse brain. Nature 445, 168–176 (2007). [DOI] [PubMed] [Google Scholar]
35.Dorr A., Sled J. G., Kabani N., Three-dimensional cerebral vasculature of the CBA mouse brain: A magnetic resonance imaging and micro computed tomography study. Neuroimage 35, 1409–1423 (2007). [DOI] [PubMed] [Google Scholar]
36.Xiong B., Li A., Lou Y., Chen S., Long B., Peng J., Yang Z., Xu T., Yang X., Li X., Jiang T., Luo Q., Gong H., Precise cerebral vascular atlas in stereotaxic coordinates of whole mouse brain. Front. Neuroanat 11, 128 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
37.Turner R., Bihan D. L., Moonen C. T., Despres D., Frank J., Echo-planar time course MRI of cat brain oxygenation changes. Magn. Reson. Med. 22, 159–166 (1991). [DOI] [PubMed] [Google Scholar]
38.Bulte D., Chiarelli P., Wise R., Jezzard P., Measurement of cerebral blood volume in humans using hyperoxic MRI contrast. J. Magn. Reson. Imaging 26, 894–899 (2007). [DOI] [PubMed] [Google Scholar]
39.MacDonald M. E., Berman A. J. L., Mazerolle E. L., Williams R. J., Pike G. B., Modeling hyperoxia-induced BOLD signal dynamics to estimate cerebral blood flow, volume and mean transit time. Neuroimage 178, 461–474 (2018). [DOI] [PubMed] [Google Scholar]
40.Vu C., Xu B., González-Zacarías C., Shen J., Baas K. P. A., Choi S., Nederveen A. J., Wood J. C., Sinusoidal CO₂ respiratory challenge for concurrent perfusion and cerebrovascular reactivity MRI. Front. Physiol. 14, 1102983 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
41.Kety S. S., Schmidt C. F., The effects of altered arterial tensions of carbon dioxide and oxygen on cerebral blood flow and cerebral oxygen consumption of normal young men. J. Clin. Invest. 27, 484–492 (1948). [DOI] [PMC free article] [PubMed] [Google Scholar]
42.Kim S. G., Rostrup E., Larsson H. B., Ogawa S., Paulson O. B., Determination of relative CMRO₂ from CBF and BOLD changes: Significant increase of oxygen consumption rate during visual stimulation. Magn. Reson. Med. 41, 1152–1161 (1999). [DOI] [PubMed] [Google Scholar]
43.Davis T. L., Kwong K. K., Weisskoff R. M., Rosen B. R., Calibrated functional MRI: Mapping the dynamics of oxidative metabolism. Proc. Natl. Acad. Sci. 95, 1834–1839 (1998). [DOI] [PMC free article] [PubMed] [Google Scholar]
44.Hoge R. D., Atkinson J., Gill B., Crelier G. R., Marrett S., Pike G. B., Linear coupling between cerebral blood flow and oxygen consumption in activated human cortex. Proc. Natl. Acad. Sci. 96, 9403–9408 (1999). [DOI] [PMC free article] [PubMed] [Google Scholar]
45.Parkes L. M., Rashid W., Chard D. T., Tofts P. S., Normal cerebral perfusion measurements using arterial spin labeling: Reproducibility, stability, and age and gender effects. Magn. Reson. Med. 51, 736–743 (2004). [DOI] [PubMed] [Google Scholar]
46.Jahng G.-H., Song E., Zhu X.-P., Matson G. B., Weiner M. W., Schuff N., Human brain: Reliability and reproducibility of pulsed arterial spin-labeling perfusion MR imaging. Radiology 234, 909–916 (2005). [DOI] [PMC free article] [PubMed] [Google Scholar]
47.Thulborn K. R., Waterton J. C., Matthews P. M., Radda G. K., Oxygenation dependence of the transverse relaxation time of water protons in whole blood at high field. Biochim. Biophys. Acta 714, 265–270 (1982). [DOI] [PubMed] [Google Scholar]
48.Brown E. N., Purdon P. L., Van Dort C. J., General anesthesia and altered states of arousal: A systems neuroscience analysis. Annu. Rev. Neurosci. 34, 601–628 (2011). [DOI] [PMC free article] [PubMed] [Google Scholar]
49.Kohrs R., Durieux M. E., Ketamine: Teaching an old drug new tricks. Anesth. Analg. 87, 1186–1193 (1998). [DOI] [PubMed] [Google Scholar]
50.Hentschke H., Schwarz C., Antkowiak B., Neocortex is the major target of sedative concentrations of volatile anaesthetics: Strong depression of firing rates and increase of GABA_A receptor-mediated inhibition. Eur. J. Neurosci. 21, 93–102 (2005). [DOI] [PubMed] [Google Scholar]
51.Masamoto K., Fukuda M., Vazquez A., Kim S. G., Dose-dependent effect of isoflurane on neurovascular coupling in rat cerebral cortex. Eur. J. Neurosci. 30, 242–250 (2009). [DOI] [PMC free article] [PubMed] [Google Scholar]
52.Grandjean J., Canella C., Anckaerts C., Ayrancı G., Bougacha S., Bienert T., Buehlmann D., Coletta L., Gallino D., Gass N., Garin C. M., Nadkarni N. A., Hübner N. S., Karatas M., Komaki Y., Kreitz S., Mandino F., Mechling A. E., Sato C., Sauer K., Shah D., Strobelt S., Takata N., Wank I., Wu T., Yahata N., Yeow L. Y., Yee Y., Aoki I., Chakravarty M. M., Chang W.-T., Dhenain M., von Elverfeldt D., Harsan L.-A., Hess A., Jiang T., Keliris G. A., Lerch J. P., Meyer-Lindenberg A., Okano H., Rudin M., Sartorius A., Van der Linden A., Verhoye M., Weber-Fahr W., Wenderoth N., Zerbi V., Gozzi A., Common functional networks in the mouse brain revealed by multi-centre resting-state fMRI analysis. Neuroimage 205, 116278 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
53.Schroeter A., Schlegel F., Seuwen A., Grandjean J., Rudin M., Specificity of stimulus-evoked fMRI responses in the mouse: The influence of systemic physiological changes associated with innocuous stimulation under four different anesthetics. Neuroimage 94, 372–384 (2014). [DOI] [PubMed] [Google Scholar]
54.Shim H.-J., Jung W. B., Schlegel F., Lee J., Kim S., Lee J., Kim S.-G., Mouse fMRI under ketamine and xylazine anesthesia: Robust contralateral somatosensory cortex activation in response to forepaw stimulation. Neuroimage 177, 30–44 (2018). [DOI] [PubMed] [Google Scholar]
55.Zhao F., Meng X., Lu S., Hyde L. A., Kennedy M. E., Houghton A. K., Evelhoch J. L., Hines C. D. G., fMRI study of olfactory processing in mice under three anesthesia protocols: Insight into the effect of ketamine on olfactory processing. Neuroimage 213, 116725 (2020). [DOI] [PubMed] [Google Scholar]
56.Sicard K., Shen Q., Brevard M. E., Sullivan R., Ferris C. F., King J. A., Duong T. Q., Regional cerebral blood flow and BOLD responses in conscious and anesthetized rats under basal and hypercapnic conditions: Implications for functional MRI studies. J. Cereb. Blood Flow Metab. 23, 472–481 (2003). [DOI] [PMC free article] [PubMed] [Google Scholar]
57.Matta B. F., Heath K. J., Tipping K., Summors A. C., Direct cerebral vasodilatory effects of sevoflurane and isoflurane. Anesthesiology 91, 677–680 (1999). [DOI] [PubMed] [Google Scholar]
58.Moens Y., Fargetton X., A comparative study of medetomidine/ketamine and xylazine/ketamine anaesthesia in dogs. Vet. Rec. 127, 567–571 (1990). [PubMed] [Google Scholar]
59.Difilippo S. M., Norberg P. J., Suson U. D., Savino A. M., Reim D. A., A comparison of xylazine and medetomidine in an anesthetic combination in New Zealand White Rabbits. Contemp. Top. Lab. Anim. Sci. 43, 32–34 (2004). [PubMed] [Google Scholar]
60.You T., Im G. H., Kim S. G., Characterization of brain-wide somatosensory BOLD fMRI in mice under dexmedetomidine/isoflurane and ketamine/xylazine. Sci. Rep. 11, 13110 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
61.Zheng B., Lee P. T. H., Golay X., High-sensitivity cerebral perfusion mapping in mice by kbGRASE-FAIR at 9.4 T. NMR Biomed. 23, 1061–1070 (2010). [DOI] [PubMed] [Google Scholar]
62.Sullender C. T., Richards L. M., He F., Luan L., Dunn A. K., Dynamics of isoflurane-induced vasodilation and blood flow of cerebral vasculature revealed by multi-exposure speckle imaging. J. Neurosci. Methods 366, 109434 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
63.Takuwa H., Tajima Y., Kokuryo D., Matsuura T., Kawaguchi H., Masamoto K., Taniguchi J., Ikoma Y., Seki C., Aoki I., Tomita Y., Suzuki N., Kanno I., Ito H., Hemodynamic changes during neural deactivation in awake mice: A measurement by laser-Doppler flowmetry in crossed cerebellar diaschisis. Brain Res. 1537, 350–355 (2013). [DOI] [PubMed] [Google Scholar]
64.Willats L., Calamante F., The 39 steps: Evading error and deciphering the secrets for accurate dynamic susceptibility contrast MRI. NMR Biomed. 26, 913–931 (2013). [DOI] [PubMed] [Google Scholar]
65.Knutsson L., Ståhlberg F., Wirestam R., Absolute quantification of perfusion using dynamic susceptibility contrast MRI: Pitfalls and possibilities. MAGMA 23, 1–21 (2010). [DOI] [PubMed] [Google Scholar]
66.Calamante F., Perfusion MRI using dynamic-susceptibility contrast MRI: Quantification issues in patient studies. Top. Magn. Reson. Imaging 21, 75–85 (2010). [DOI] [PubMed] [Google Scholar]
67.Kim S.-G., Ogawa S., Biophysical and physiological origins of blood oxygenation level-dependent fMRI signals. J. Cereb. Blood Flow Metab. 32, 1188–1206 (2012). [DOI] [PMC free article] [PubMed] [Google Scholar]
68.Schulman J. B., Sayin E. S., Manalac A., Poublanc J., Sobczyk O., Duffin J., Fisher J. A., Mikulis D., Uludağ K., DSC MRI in the human brain using deoxyhemoglobin and gadolinium–Simulations and validations at 3T. Front. Neuroimaging 2, 1048652 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
69.Lee S.-P., Silva A. C., Ugurbil K., Kim S.-G., Diffusion-weighted spin-echo fMRI at 9.4 T: Microvascular/tissue contribution to BOLD signal changes. Magn. Reson. Med. 42, 919–928 (1999). [DOI] [PubMed] [Google Scholar]
70.Hernandez D., Kim K.-N., A review on the RF coil designs and trends for ultra high field magnetic resonance imaging. Invest. Magn. Reson. Imaging 24, 95–122 (2020). [Google Scholar]
71.Lee H.-L., Li Z., Coulson E. J., Chuang K.-H., Ultrafast fMRI of the rodent brain using simultaneous multi-slice EPI. Neuroimage 195, 48–58 (2019). [DOI] [PubMed] [Google Scholar]
72.Lustig M., Donoho D. L., Santos J. M., Pauly J. M., Compressed sensing MRI. IEEE Signal Process. Mag. 25, 72–82 (2008). [Google Scholar]
73.Hollingsworth K. G., Reducing acquisition time in clinical MRI by data undersampling and compressed sensing reconstruction. Phys. Med. Biol. 60, R297–R322 (2015). [DOI] [PubMed] [Google Scholar]
74.Mazze R. I., Rice S. A., Baden J. M., Halothane, isoflurane, and enflurane MAC in pregnant and nonpregnant female and male mice and rats. Anesthesiology 62, 339–341 (1985). [DOI] [PubMed] [Google Scholar]
75.Lin C. Y., Uptake of anaesthetic gases and vapours. Anaesth. Intensive Care 22, 363–373 (1994). [DOI] [PubMed] [Google Scholar]
76.Antognini J. F., Wang X. W., Carstens E., Quantitative and qualitative effects of isoflurane on movement occurring after noxious stimulation. Anesthesiology 91, 1064–1064 (1999). [DOI] [PubMed] [Google Scholar]
77.Shen T., Weissleder R., Papisov M., Bogdanov A. Jr., Brady T. J., Monocrystalline iron oxide nanocompounds (MION): Physicochemical properties. Magn. Reson. Med. 29, 599–604 (1993). [DOI] [PubMed] [Google Scholar]
78.Cox R. W., AFNI: Software for analysis and visualization of functional magnetic resonance neuroimages. Comput. Biomed. Res. 29, 162–173 (1996). [DOI] [PubMed] [Google Scholar]
79.Østergaard L., Weisskoff R. M., Chesler D. A., Gyldensted C., Rosen B. R., High resolution measurement of cerebral blood flow using intravascular tracer bolus passages. Part I: Mathematical approach and statistical analysis. Magn. Reson. Med. 36, 715–725 (1996). [DOI] [PubMed] [Google Scholar]
80.Villringer A., Rosen B. R., Belliveau J. W., Ackerman J. L., Lauffer R. B., Buxton R. B., Chao Y. S., Wedeenand V. J., Brady T. J., Dynamic imaging with lanthanide chelates in normal brain: Contrast due to magnetic susceptibility effects. Magn. Reson. Med. 6, 164–174 (1988). [DOI] [PubMed] [Google Scholar]
81.Rosen B. R., Belliveau J. W., Vevea J. M., Brady T. J., Perfusion imaging with NMR contrast agents. Magn. Reson. Med. 14, 249–265 (1990). [DOI] [PubMed] [Google Scholar]
82.Simonsen C. Z., Østergaard L., Vestergaard-Poulsen P., Røhl L., Bjørnerud A., Gyldensted C., CBF and CBV measurements by USPIO bolus tracking: Reproducibility and comparison with Gd-based values. J. Magn. Reson. Imaging 9, 342–347 (1999). [DOI] [PubMed] [Google Scholar]
83.Peruzzo D., Bertoldo A., Zanderigo F., Cobelli C., Automatic selection of arterial input function on dynamic contrast-enhanced MR images. Comput. Methods Programs Biomed. 104, e148–e157 (2011). [DOI] [PubMed] [Google Scholar]
84.Calamante F., Thomas D. L., Pell G. S., Wiersma J., Turner R., Measuring cerebral blood flow using magnetic resonance imaging techniques. J. Cereb. Blood Flow Metab. 19, 701–735 (1999). [DOI] [PubMed] [Google Scholar]
85.Wu O., Østergaard L., Weisskoff R. M., Benner T., Rosen B. R., Sorensen A. G., Tracer arrival timing-insensitive technique for estimating flow in MR perfusion-weighted imaging using singular value decomposition with a block-circulant deconvolution matrix. Magn. Reson. Med. 50, 164–174 (2003). [DOI] [PubMed] [Google Scholar]
86.Wirestam R., Andersson L., Ostergaard L., Bolling M., Aunola J. P., Lindgren A., Geijer B., Holtås S., Ståhlberg F., Assessment of regional cerebral blood flow by dynamic susceptibility contrast MRI using different deconvolution techniques. Magn. Reson. Med. 43, 691–700 (2000). [DOI] [PubMed] [Google Scholar]
87.Østergaard L., Principles of cerebral perfusion imaging by bolus tracking. J. Magn. Reson. Imaging 22, 710–717 (2005). [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Figs. S1 to S8

Tables S1 to S3

sciadv.adm7605_sm.pdf^{(2.2MB, pdf)}

[R1] 1.L. Edvinsson, E. T. MacKenzie, J. McCulloch, Cerebral Blood Flow and Metabolism (Raven Press, 1993), vol. 1185. [Google Scholar]

[R2] 2.Wintermark M., Sesay M., Barbier E., Borbély K., Dillon W. P., Eastwood J. D., Glenn T. C., Grandin C. B., Pedraza S., Soustiel J.-F., Nariai T., Zaharchuk G., Caillé J.-M., Dousset V., Yonas H., Comparative overview of brain perfusion imaging techniques. Stroke 36, e83–e99 (2005). [DOI] [PubMed] [Google Scholar]

[R3] 3.Vu C., Chai Y., Coloigner J., Nederveen A. J., Borzage M., Bush A., Wood J. C., Quantitative perfusion mapping with induced transient hypoxia using BOLD MRI. Magn. Reson. Med. 85, 168–181 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R4] 4.Lee D., Le T. T., Im G. H., Kim S.-G., Whole-brain perfusion mapping in mice by dynamic BOLD MRI with transient hypoxia. J. Cereb. Blood Flow Metab. 42, 2270–2286 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R5] 5.Poublanc J., Sobczyk O., Shafi R., Sayin E. S., Schulman J., Duffin J., Uludag K., Wood J. C., Vu C., Dharmakumar R., Fisher J. A., Mikulis D. J., Perfusion MRI using endogenous deoxyhemoglobin as a contrast agent: Preliminary data. Magn. Reson. Med. 86, 3012–3021 (2021). [DOI] [PubMed] [Google Scholar]

[R6] 6.Cryan J. F., Holmes A., The ascent of mouse: Advances in modelling human depression and anxiety. Nat. Rev. Drug Discov. 4, 775–790 (2005). [DOI] [PubMed] [Google Scholar]

[R7] 7.Götz J., Ittner L. M., Animal models of Alzheimer's disease and frontotemporal dementia. Nat. Rev. Neurosci. 9, 532–544 (2008). [DOI] [PubMed] [Google Scholar]

[R8] 8.Nestler E. J., Hyman S. E., Animal models of neuropsychiatric disorders. Nat. Neurosci. 13, 1161–1169 (2010). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R9] 9.Gehrmann J., Hammer P. E., Maguire C. T., Wakimoto H., Triedman J. K., Berul C. I., Phenotypic screening for heart rate variability in the mouse. Am. J. Physiol. Heart Circ. Physiol. 279, H733–H740 (2000). [DOI] [PubMed] [Google Scholar]

[R10] 10.Janssen B. J., De Celle T., Debets J. J., Brouns A. E., Callahan M. F., Smith T. L., Effects of anesthetics on systemic hemodynamics in mice. Am. J. Physiol. Heart Circ. Physiol. 287, H1618–H1624 (2004). [DOI] [PubMed] [Google Scholar]

[R11] 11.Szczęsny G., Veihelmann A., Massberg S., Nolte D., Messmer K., Long-term anaesthesia using inhalatory isoflurane in different strains of mice—The haemodynamic effects. Lab. Anim 38, 64–69 (2004). [DOI] [PubMed] [Google Scholar]

[R12] 12.Iida H., Ohata H., Iida M., Watanabe Y., Dohi S., Isoflurane and sevoflurane induce vasodilation of cerebral vessels via ATP-sensitive K⁺channel activation. Anesthesiology 89, 954–960 (1998). [DOI] [PubMed] [Google Scholar]

[R13] 13.Masamoto K., Kanno I., Anesthesia and the quantitative evaluation of neurovascular coupling. J. Cereb. Blood Flow Metab. 32, 1233–1247 (2012). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R14] 14.Weber R., Ramos-Cabrer P., Wiedermann D., Van Camp N., Hoehn M., A fully noninvasive and robust experimental protocol for longitudinal fMRI studies in the rat. Neuroimage 29, 1303–1310 (2006). [DOI] [PubMed] [Google Scholar]

[R15] 15.Fukuda M., Vazquez A. L., Zong X., Kim S. G., Effects of the α2-adrenergic receptor agonist dexmedetomidine on neural, vascular and BOLD fMRI responses in the somatosensory cortex. Eur. J. Neurosci. 37, 80–95 (2013). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R16] 16.Grandjean J., Schroeter A., Batata I., Rudin M., Optimization of anesthesia protocol for resting-state fMRI in mice based on differential effects of anesthetics on functional connectivity patterns. Neuroimage 102, 838–847 (2014). [DOI] [PubMed] [Google Scholar]

[R17] 17.Adamczak J. M., Farr T. D., Seehafer J. U., Kalthoff D., Hoehn M., High field BOLD response to forepaw stimulation in the mouse. Neuroimage 51, 704–712 (2010). [DOI] [PubMed] [Google Scholar]

[R18] 18.Sinclair M. D., A review of the physiological effects of α₂-agonists related to the clinical use of medetomidine in small animal practice. Can. Vet. J. 44, 885–897 (2003). [PMC free article] [PubMed] [Google Scholar]

[R19] 19.Munting L. P., Derieppe M. P., Suidgeest E., Denis de Senneville B., Wells J. A., van der Weerd L., Influence of different isoflurane anesthesia protocols on murine cerebral hemodynamics measured with pseudo-continuous arterial spin labeling. NMR Biomed. 32, e4105 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R20] 20.Munting L. P., Derieppe M., Suidgeest E., Hirschler L., van Osch M. J., de Senneville B. D., van der Weerd L., Cerebral blood flow and cerebrovascular reactivity are preserved in a mouse model of cerebral microvascular amyloidosis. eLife 10, e61279 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R21] 21.Muir E. R., Shen Q., Duong T. Q., Cerebral blood flow MRI in mice using the cardiac-spin-labeling technique. Magn. Reson. Med. 60, 744–748 (2008). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R22] 22.Smith M. R., Lu H., Frayne R., Signal-to-noise ratio effects in quantitative cerebral perfusion using dynamic susceptibility contrast agents. Magn. Reson. Med. 49, 122–128 (2003). [DOI] [PubMed] [Google Scholar]

[R23] 23.Knutsson L., Ståhlberg F., Wirestam R., Aspects on the accuracy of cerebral perfusion parameters obtained by dynamic susceptibility contrast MRI: A simulation study. Magn. Reson. Imaging 22, 789–798 (2004). [DOI] [PubMed] [Google Scholar]

[R24] 24.Ellinger R., Kremser C., Schocke M. F., Kolbitsch C., Griebel J., Felber S. R., Aichner F. T.et al. , J. Comput. Assist. Tomogr. 24, 942–948 (2000). [DOI] [PubMed] [Google Scholar]

[R25] 25.Han S., Son J. P., Cho H., Park J. Y., Kim S. G., Gradient-echo and spin-echo blood oxygenation level–dependent functional MRI at ultrahigh fields of 9.4 and 15.2 Tesla. Magn. Reson. Med. 81, 1237–1246 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R26] 26.Knutsson L., Börjesson S., Larsson E. M., Risberg J., Gustafson L., Passant U., Ståhlberg F., Wirestam R., Absolute quantification of cerebral blood flow in normal volunteers: Correlation between Xe-133 SPECT and dynamic susceptibility contrast MRI. J. Magn. Reson. Imaging 26, 913–920 (2007). [DOI] [PubMed] [Google Scholar]

[R27] 27.Knutsson L., Lindgren E., Ahlgren A., van Osch M. J. P., Bloch K. M., Surova Y., Ståhlberg F., Van Westen D., Wirestam R., Dynamic susceptibility contrast MRI with a prebolus contrast agent administration design for improved absolute quantification of perfusion. Magn. Reson. Med. 72, 996–1006 (2014). [DOI] [PubMed] [Google Scholar]

[R28] 28.Moreno H., Hua F., Brown T., Small S., Longitudinal mapping of mouse cerebral blood volume with MRI. NMR Biomed. 19, 535–543 (2006). [DOI] [PubMed] [Google Scholar]

[R29] 29.Kim S. G., Harel N., Jin T., Kim T., Lee P., Zhao F., Cerebral blood volume MRI with intravascular superparamagnetic iron oxide nanoparticles. NMR Biomed. 26, 949–962 (2013). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R30] 30.Wu E. X., Wong K. K., Andrassy M., Tang H., High-resolution in vivo CBV mapping with MRI in wild-type mice. Magn. Reson. Med. 49, 765–770 (2003). [DOI] [PubMed] [Google Scholar]

[R31] 31.Ogawa S., Menon R., Tank D. W., Kim S., Merkle H., Ellermann J., Ugurbil K., Functional brain mapping by blood oxygenation level-dependent contrast magnetic resonance imaging. A comparison of signal characteristics with a biophysical model. Biophys. J. 64, 803–812 (1993). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R32] 32.Boxerman J. L., Hamberg L. M., Rosen B. R., Weisskoff R. M., MR contrast due to intravascular magnetic susceptibility perturbations. Magn. Reson. Med. 34, 555–566 (1995). [DOI] [PubMed] [Google Scholar]

[R33] 33.Uludağ K., Müller-Bierl B., Uğurbil K., An integrative model for neuronal activity-induced signal changes for gradient and spin echo functional imaging. Neuroimage 48, 150–165 (2009). [DOI] [PubMed] [Google Scholar]

[R34] 34.Lein E. S., Hawrylycz M. J., Ao N., Ayres M., Bensinger A., Bernard A., Boe A. F., Boguski M. S., Brockway K. S., Byrnes E. J., Chen L., Chen L., Chen T.-M., Chin M. C., Chong J., Crook B. E., Czaplinska A., Dang C. N., Datta S., Dee N. R., Desaki A. L., Desta T., Diep E., Dolbeare T. A., Donelan M. J., Dong H.-W., Dougherty J. G., Duncan B. J., Ebbert A. J., Eichele G., Estin L. K., Faber C., Facer B. A., Fields R., Fischer S. R., Fliss T. P., Frensley C., Gates S. N., Glattfelder K. J., Halverson K. R., Hart M. R., Hohmann J. G., Howell M. P., Jeung D. P., Johnson R. A., Karr P. T., Kawal R., Kidney J. M., Knapik R. H., Kuan C. L., Lake J. H., Laramee A. R., Larsen K. D., Lau C., Lemon T. A., Liang A. J., Liu Y., Luong L. T., Michaels J., Morgan J. J., Morgan R. J., Mortrud M. T., Mosqueda N. F., Ng L. L., Ng R., Orta G. J., Overly C. C., Pak T. H., Parry S. E., Pathak S. D., Pearson O. C., Puchalski R. B., Riley Z. L., Rockett H. R., Rowland S. A., Royall J. J., Ruiz M. J., Sarno N. R., Schaffnit K., Shapovalova N. V., Sivisay T., Slaughterbeck C. R., Smith S. C., Smith K. A., Smith B. I., Sodt A. J., Stewart N. N., Stumpf K.-R., Sunkin S. M., Sutram M., Tam A., Teemer C. D., Thaller C., Thompson C. L., Varnam L. R., Visel A., Whitlock R. M., Wohnoutka P. E., Wolkey C. K., Wong V. Y., Wood M., Yaylaoglu M. B., Young R. C., Youngstrom B. L., Yuan X. F., Zhang B., Zwingman T. A., Jones A. R., Genome-wide atlas of gene expression in the adult mouse brain. Nature 445, 168–176 (2007). [DOI] [PubMed] [Google Scholar]

[R35] 35.Dorr A., Sled J. G., Kabani N., Three-dimensional cerebral vasculature of the CBA mouse brain: A magnetic resonance imaging and micro computed tomography study. Neuroimage 35, 1409–1423 (2007). [DOI] [PubMed] [Google Scholar]

[R36] 36.Xiong B., Li A., Lou Y., Chen S., Long B., Peng J., Yang Z., Xu T., Yang X., Li X., Jiang T., Luo Q., Gong H., Precise cerebral vascular atlas in stereotaxic coordinates of whole mouse brain. Front. Neuroanat 11, 128 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R37] 37.Turner R., Bihan D. L., Moonen C. T., Despres D., Frank J., Echo-planar time course MRI of cat brain oxygenation changes. Magn. Reson. Med. 22, 159–166 (1991). [DOI] [PubMed] [Google Scholar]

[R38] 38.Bulte D., Chiarelli P., Wise R., Jezzard P., Measurement of cerebral blood volume in humans using hyperoxic MRI contrast. J. Magn. Reson. Imaging 26, 894–899 (2007). [DOI] [PubMed] [Google Scholar]

[R39] 39.MacDonald M. E., Berman A. J. L., Mazerolle E. L., Williams R. J., Pike G. B., Modeling hyperoxia-induced BOLD signal dynamics to estimate cerebral blood flow, volume and mean transit time. Neuroimage 178, 461–474 (2018). [DOI] [PubMed] [Google Scholar]

[R40] 40.Vu C., Xu B., González-Zacarías C., Shen J., Baas K. P. A., Choi S., Nederveen A. J., Wood J. C., Sinusoidal CO₂ respiratory challenge for concurrent perfusion and cerebrovascular reactivity MRI. Front. Physiol. 14, 1102983 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R41] 41.Kety S. S., Schmidt C. F., The effects of altered arterial tensions of carbon dioxide and oxygen on cerebral blood flow and cerebral oxygen consumption of normal young men. J. Clin. Invest. 27, 484–492 (1948). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R42] 42.Kim S. G., Rostrup E., Larsson H. B., Ogawa S., Paulson O. B., Determination of relative CMRO₂ from CBF and BOLD changes: Significant increase of oxygen consumption rate during visual stimulation. Magn. Reson. Med. 41, 1152–1161 (1999). [DOI] [PubMed] [Google Scholar]

[R43] 43.Davis T. L., Kwong K. K., Weisskoff R. M., Rosen B. R., Calibrated functional MRI: Mapping the dynamics of oxidative metabolism. Proc. Natl. Acad. Sci. 95, 1834–1839 (1998). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R44] 44.Hoge R. D., Atkinson J., Gill B., Crelier G. R., Marrett S., Pike G. B., Linear coupling between cerebral blood flow and oxygen consumption in activated human cortex. Proc. Natl. Acad. Sci. 96, 9403–9408 (1999). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R45] 45.Parkes L. M., Rashid W., Chard D. T., Tofts P. S., Normal cerebral perfusion measurements using arterial spin labeling: Reproducibility, stability, and age and gender effects. Magn. Reson. Med. 51, 736–743 (2004). [DOI] [PubMed] [Google Scholar]

[R46] 46.Jahng G.-H., Song E., Zhu X.-P., Matson G. B., Weiner M. W., Schuff N., Human brain: Reliability and reproducibility of pulsed arterial spin-labeling perfusion MR imaging. Radiology 234, 909–916 (2005). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R47] 47.Thulborn K. R., Waterton J. C., Matthews P. M., Radda G. K., Oxygenation dependence of the transverse relaxation time of water protons in whole blood at high field. Biochim. Biophys. Acta 714, 265–270 (1982). [DOI] [PubMed] [Google Scholar]

[R48] 48.Brown E. N., Purdon P. L., Van Dort C. J., General anesthesia and altered states of arousal: A systems neuroscience analysis. Annu. Rev. Neurosci. 34, 601–628 (2011). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R49] 49.Kohrs R., Durieux M. E., Ketamine: Teaching an old drug new tricks. Anesth. Analg. 87, 1186–1193 (1998). [DOI] [PubMed] [Google Scholar]

[R50] 50.Hentschke H., Schwarz C., Antkowiak B., Neocortex is the major target of sedative concentrations of volatile anaesthetics: Strong depression of firing rates and increase of GABA_A receptor-mediated inhibition. Eur. J. Neurosci. 21, 93–102 (2005). [DOI] [PubMed] [Google Scholar]

[R51] 51.Masamoto K., Fukuda M., Vazquez A., Kim S. G., Dose-dependent effect of isoflurane on neurovascular coupling in rat cerebral cortex. Eur. J. Neurosci. 30, 242–250 (2009). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R52] 52.Grandjean J., Canella C., Anckaerts C., Ayrancı G., Bougacha S., Bienert T., Buehlmann D., Coletta L., Gallino D., Gass N., Garin C. M., Nadkarni N. A., Hübner N. S., Karatas M., Komaki Y., Kreitz S., Mandino F., Mechling A. E., Sato C., Sauer K., Shah D., Strobelt S., Takata N., Wank I., Wu T., Yahata N., Yeow L. Y., Yee Y., Aoki I., Chakravarty M. M., Chang W.-T., Dhenain M., von Elverfeldt D., Harsan L.-A., Hess A., Jiang T., Keliris G. A., Lerch J. P., Meyer-Lindenberg A., Okano H., Rudin M., Sartorius A., Van der Linden A., Verhoye M., Weber-Fahr W., Wenderoth N., Zerbi V., Gozzi A., Common functional networks in the mouse brain revealed by multi-centre resting-state fMRI analysis. Neuroimage 205, 116278 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R53] 53.Schroeter A., Schlegel F., Seuwen A., Grandjean J., Rudin M., Specificity of stimulus-evoked fMRI responses in the mouse: The influence of systemic physiological changes associated with innocuous stimulation under four different anesthetics. Neuroimage 94, 372–384 (2014). [DOI] [PubMed] [Google Scholar]

[R54] 54.Shim H.-J., Jung W. B., Schlegel F., Lee J., Kim S., Lee J., Kim S.-G., Mouse fMRI under ketamine and xylazine anesthesia: Robust contralateral somatosensory cortex activation in response to forepaw stimulation. Neuroimage 177, 30–44 (2018). [DOI] [PubMed] [Google Scholar]

[R55] 55.Zhao F., Meng X., Lu S., Hyde L. A., Kennedy M. E., Houghton A. K., Evelhoch J. L., Hines C. D. G., fMRI study of olfactory processing in mice under three anesthesia protocols: Insight into the effect of ketamine on olfactory processing. Neuroimage 213, 116725 (2020). [DOI] [PubMed] [Google Scholar]

[R56] 56.Sicard K., Shen Q., Brevard M. E., Sullivan R., Ferris C. F., King J. A., Duong T. Q., Regional cerebral blood flow and BOLD responses in conscious and anesthetized rats under basal and hypercapnic conditions: Implications for functional MRI studies. J. Cereb. Blood Flow Metab. 23, 472–481 (2003). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R57] 57.Matta B. F., Heath K. J., Tipping K., Summors A. C., Direct cerebral vasodilatory effects of sevoflurane and isoflurane. Anesthesiology 91, 677–680 (1999). [DOI] [PubMed] [Google Scholar]

[R58] 58.Moens Y., Fargetton X., A comparative study of medetomidine/ketamine and xylazine/ketamine anaesthesia in dogs. Vet. Rec. 127, 567–571 (1990). [PubMed] [Google Scholar]

[R59] 59.Difilippo S. M., Norberg P. J., Suson U. D., Savino A. M., Reim D. A., A comparison of xylazine and medetomidine in an anesthetic combination in New Zealand White Rabbits. Contemp. Top. Lab. Anim. Sci. 43, 32–34 (2004). [PubMed] [Google Scholar]

[R60] 60.You T., Im G. H., Kim S. G., Characterization of brain-wide somatosensory BOLD fMRI in mice under dexmedetomidine/isoflurane and ketamine/xylazine. Sci. Rep. 11, 13110 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R61] 61.Zheng B., Lee P. T. H., Golay X., High-sensitivity cerebral perfusion mapping in mice by kbGRASE-FAIR at 9.4 T. NMR Biomed. 23, 1061–1070 (2010). [DOI] [PubMed] [Google Scholar]

[R62] 62.Sullender C. T., Richards L. M., He F., Luan L., Dunn A. K., Dynamics of isoflurane-induced vasodilation and blood flow of cerebral vasculature revealed by multi-exposure speckle imaging. J. Neurosci. Methods 366, 109434 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R63] 63.Takuwa H., Tajima Y., Kokuryo D., Matsuura T., Kawaguchi H., Masamoto K., Taniguchi J., Ikoma Y., Seki C., Aoki I., Tomita Y., Suzuki N., Kanno I., Ito H., Hemodynamic changes during neural deactivation in awake mice: A measurement by laser-Doppler flowmetry in crossed cerebellar diaschisis. Brain Res. 1537, 350–355 (2013). [DOI] [PubMed] [Google Scholar]

[R64] 64.Willats L., Calamante F., The 39 steps: Evading error and deciphering the secrets for accurate dynamic susceptibility contrast MRI. NMR Biomed. 26, 913–931 (2013). [DOI] [PubMed] [Google Scholar]

[R65] 65.Knutsson L., Ståhlberg F., Wirestam R., Absolute quantification of perfusion using dynamic susceptibility contrast MRI: Pitfalls and possibilities. MAGMA 23, 1–21 (2010). [DOI] [PubMed] [Google Scholar]

[R66] 66.Calamante F., Perfusion MRI using dynamic-susceptibility contrast MRI: Quantification issues in patient studies. Top. Magn. Reson. Imaging 21, 75–85 (2010). [DOI] [PubMed] [Google Scholar]

[R67] 67.Kim S.-G., Ogawa S., Biophysical and physiological origins of blood oxygenation level-dependent fMRI signals. J. Cereb. Blood Flow Metab. 32, 1188–1206 (2012). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R68] 68.Schulman J. B., Sayin E. S., Manalac A., Poublanc J., Sobczyk O., Duffin J., Fisher J. A., Mikulis D., Uludağ K., DSC MRI in the human brain using deoxyhemoglobin and gadolinium–Simulations and validations at 3T. Front. Neuroimaging 2, 1048652 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R69] 69.Lee S.-P., Silva A. C., Ugurbil K., Kim S.-G., Diffusion-weighted spin-echo fMRI at 9.4 T: Microvascular/tissue contribution to BOLD signal changes. Magn. Reson. Med. 42, 919–928 (1999). [DOI] [PubMed] [Google Scholar]

[R70] 70.Hernandez D., Kim K.-N., A review on the RF coil designs and trends for ultra high field magnetic resonance imaging. Invest. Magn. Reson. Imaging 24, 95–122 (2020). [Google Scholar]

[R71] 71.Lee H.-L., Li Z., Coulson E. J., Chuang K.-H., Ultrafast fMRI of the rodent brain using simultaneous multi-slice EPI. Neuroimage 195, 48–58 (2019). [DOI] [PubMed] [Google Scholar]

[R72] 72.Lustig M., Donoho D. L., Santos J. M., Pauly J. M., Compressed sensing MRI. IEEE Signal Process. Mag. 25, 72–82 (2008). [Google Scholar]

[R73] 73.Hollingsworth K. G., Reducing acquisition time in clinical MRI by data undersampling and compressed sensing reconstruction. Phys. Med. Biol. 60, R297–R322 (2015). [DOI] [PubMed] [Google Scholar]

[R74] 74.Mazze R. I., Rice S. A., Baden J. M., Halothane, isoflurane, and enflurane MAC in pregnant and nonpregnant female and male mice and rats. Anesthesiology 62, 339–341 (1985). [DOI] [PubMed] [Google Scholar]

[R75] 75.Lin C. Y., Uptake of anaesthetic gases and vapours. Anaesth. Intensive Care 22, 363–373 (1994). [DOI] [PubMed] [Google Scholar]

[R76] 76.Antognini J. F., Wang X. W., Carstens E., Quantitative and qualitative effects of isoflurane on movement occurring after noxious stimulation. Anesthesiology 91, 1064–1064 (1999). [DOI] [PubMed] [Google Scholar]

[R77] 77.Shen T., Weissleder R., Papisov M., Bogdanov A. Jr., Brady T. J., Monocrystalline iron oxide nanocompounds (MION): Physicochemical properties. Magn. Reson. Med. 29, 599–604 (1993). [DOI] [PubMed] [Google Scholar]

[R78] 78.Cox R. W., AFNI: Software for analysis and visualization of functional magnetic resonance neuroimages. Comput. Biomed. Res. 29, 162–173 (1996). [DOI] [PubMed] [Google Scholar]

[R79] 79.Østergaard L., Weisskoff R. M., Chesler D. A., Gyldensted C., Rosen B. R., High resolution measurement of cerebral blood flow using intravascular tracer bolus passages. Part I: Mathematical approach and statistical analysis. Magn. Reson. Med. 36, 715–725 (1996). [DOI] [PubMed] [Google Scholar]

[R80] 80.Villringer A., Rosen B. R., Belliveau J. W., Ackerman J. L., Lauffer R. B., Buxton R. B., Chao Y. S., Wedeenand V. J., Brady T. J., Dynamic imaging with lanthanide chelates in normal brain: Contrast due to magnetic susceptibility effects. Magn. Reson. Med. 6, 164–174 (1988). [DOI] [PubMed] [Google Scholar]

[R81] 81.Rosen B. R., Belliveau J. W., Vevea J. M., Brady T. J., Perfusion imaging with NMR contrast agents. Magn. Reson. Med. 14, 249–265 (1990). [DOI] [PubMed] [Google Scholar]

[R82] 82.Simonsen C. Z., Østergaard L., Vestergaard-Poulsen P., Røhl L., Bjørnerud A., Gyldensted C., CBF and CBV measurements by USPIO bolus tracking: Reproducibility and comparison with Gd-based values. J. Magn. Reson. Imaging 9, 342–347 (1999). [DOI] [PubMed] [Google Scholar]

[R83] 83.Peruzzo D., Bertoldo A., Zanderigo F., Cobelli C., Automatic selection of arterial input function on dynamic contrast-enhanced MR images. Comput. Methods Programs Biomed. 104, e148–e157 (2011). [DOI] [PubMed] [Google Scholar]

[R84] 84.Calamante F., Thomas D. L., Pell G. S., Wiersma J., Turner R., Measuring cerebral blood flow using magnetic resonance imaging techniques. J. Cereb. Blood Flow Metab. 19, 701–735 (1999). [DOI] [PubMed] [Google Scholar]

[R85] 85.Wu O., Østergaard L., Weisskoff R. M., Benner T., Rosen B. R., Sorensen A. G., Tracer arrival timing-insensitive technique for estimating flow in MR perfusion-weighted imaging using singular value decomposition with a block-circulant deconvolution matrix. Magn. Reson. Med. 50, 164–174 (2003). [DOI] [PubMed] [Google Scholar]

[R86] 86.Wirestam R., Andersson L., Ostergaard L., Bolling M., Aunola J. P., Lindgren A., Geijer B., Holtås S., Ståhlberg F., Assessment of regional cerebral blood flow by dynamic susceptibility contrast MRI using different deconvolution techniques. Magn. Reson. Med. 43, 691–700 (2000). [DOI] [PubMed] [Google Scholar]

[R87] 87.Østergaard L., Principles of cerebral perfusion imaging by bolus tracking. J. Magn. Reson. Imaging 22, 710–717 (2005). [DOI] [PubMed] [Google Scholar]

PERMALINK

Mapping cerebral perfusion in mice under various anesthesia levels using highly sensitive BOLD MRI with transient hypoxia

Thuy Thi Le

Geun Ho Im

Chan Hee Lee

Sang Han Choi

Seong-Gi Kim

Roles

Abstract

INTRODUCTION

RESULTS

Experimental setup for BOLD-DSC perfusion measurements under volatile anesthesia

Fig. 1. Experimental setup for BOLD-DSC measurements in mice under volatile anesthetics.

Optimization of echo time for BOLD-DSC MRI

Fig. 2. TE dependence of the hypoxia-induced transverse relaxation rate changes.

Sensitivity and repeatability of hypoxia-induced BOLD MRI within a single imaging session

Fig. 3. Repeatability and detectability of hypoxia-induced BOLD responses within an imaging session.

Repeatability of BOLD-DSC measurements across multiple imaging sessions

Fig. 4. Repeatability of hypoxic-induced BOLD-DSC measurements across four weekly sessions.

Comparison between CBV measured by endogenous BOLD-DSC and relative CBV with exogenous contrast agents

Fig. 5. High resemblance between CBV measured by BOLD-DSC and relative CBV using MION.

Mapping CBF and CBV with BOLD-DSC MRI under different anesthetic conditions

Fig. 6. Perfusion maps of mice under four different anesthetic conditions: DEX + ISO, 1% ISO, 1.5% ISO, and 2% ISO.

Region-specific effects of anesthesia on perfusion metrics assessed by BOLD-DSC MRI

Fig. 7. Region-specific effects of anesthesia on perfusion metrics.

DISCUSSION

Precise hypoxic challenge under volatile anesthesia

Repeatability and sensitivity of BOLD response and perfusion measurements

Investigation of the effects of commonly used anesthetics in animal research

Potential limitations

MATERIALS AND METHODS

Animal preparations

MRI acquisitions

Anesthetic regimes

Isoflurane with three different doses (1, 1.5, and 2.0% ISO)

Dexmedetomidine (with 0.3% ISO supplement) protocol (DEX ± ISO)

Setup and stimulus designs

Experiment #1: Potential physiological changes caused by the absence of ISO during hypoxic stimulation

Experiment #2: Sensitivity of BOLD responses induced by varying durations of apnea stimulus

Experiment #3: TE-dependent hypoxia-induced transverse relaxation rate change ∆R2*

Experiment #4: Repeatability of weekly BOLD-DSC measurements

Experiment #5: Comparison between CBV measured by BOLD DSC and relative CBV with MION contrast agents

Experiment #6: BOLD-DSC measurements under different doses of ISO

Experiment #7: BOLD-DSC measurements under dexmedetomidine

Data processing

Quantification of perfusion metrics from arterial and tissue response curves

Suitability of the setup for BOLD-DSC with ISO (experiments #1 and #2)

TE dependence of the hypoxia-induced transverse relaxation rate changes (experiment #3)

Repeatability and detectability of hypoxic response across the whole brain within a single imaging session (experiment #5)

Repeatability of BOLD-DSC measurements across multiple imaging sessions (experiment #4)

Comparison between CBV measured by endogenous BOLD DSC and relative CBV with MION contrast agents (experiment #5)

Quantitative total and micro-vasculature perfusion metrics measured under different anesthetics (experiments #6 and #7)

Acknowledgments

Supplementary Materials

This PDF file includes:

REFERENCES AND NOTES

Associated Data

Supplementary Materials

ACTIONS

PERMALINK

RESOURCES

Similar articles

Cited by other articles

Links to NCBI Databases

Experiment #3: TE-dependent hypoxia-induced transverse relaxation rate change $∆ R_{2}^{*}$