Photoacoustic imaging detects cerebrovascular pathological changes in sepsis

Abstract

Sepsis-associated encephalopathy (SAE) is a common complication of sepsis, involving acute brain dysfunction. Although cerebrovascular impairment plays a critical role in SAE, sepsis-induced microvascular changes remain poorly quantified. Here, we used photoacoustic microscopy to dynamically assess blood-brain barrier permeability in septic mice, analyzing vascular structure across five parameters. Additionally, we examined pathological changes in major cortical regions and conducted behavioral tests to validate the findings. Results showed microvascular degeneration, including reduced vascular density and branching, alongside an increase in fine vessels. Motor-related cortical areas were most affected, correlating with severe motor and cognitive deficits in septic mice. This study provides the first in vivo, multi-parametric analysis of sepsis-induced cerebrovascular pathology, revealing region-specific damage. Our findings directly link microvascular dysfunction to SAE progression and highlight photoacoustic microscopy's potential in neuroscience research.

Keywords: Blood-brain barrier permeability; Cerebrovascular morphology; Cortical brain regions; Photoacoustic microscopy; Sepsis-associated encephalopathy.

Fig. 1

Schematic of the photoacoustic imaging system and material characterization. (a) Diagram of the 560 nm photoacoustic microscopy (PAM) system. (b) Lateral resolution of the PAM system. (c) Axial resolution of the PAM system. (d) Photographs and corresponding PAM images of blood and Evans Blue (EB) samples. (e) Comparison of PA amplitudes between blood and EB. (f) Photographs of blood and EB samples. (g) Absorption spectra of blood and EB from 400 to 700 nm, showing strong absorption at 560 nm. Data plotting and curve fitting were performed using Origin software. Abbreviations: AMP, amplifier; FWHM, full width at half maximum; DAQ, data acquisition system; FPGA, field-programmable gate array; UT, ultrasonic transducer; LSF, the line spread function; ESF, the edge spread function; abs, absorbance; a.u., arbitrary unit.

Fig. 2

Results of laser speckle blood flow imaging within 24 hours after cecal ligation and puncture (CLP) (n = 4). (a.) Laser speckle blood flow images of CLP group and Sham group within 24 hours. (b.) T-test results of laser speckle blood flow imaging between CLP mice and Sham mice at four time points (3 hours, 6 hours, 12 hours, 24 hours), with the most significant difference observed at 6 hours. (c.) Trend analysis of laser speckle blood flow imaging within 24 hours after CLP.

Fig. 3

Evaluation of inflammatory cytokine levels and liver/kidney function 6 hours after CLP modeling (n = 6). (a.) Interleukin 6 (IL-6). (b.) Tumor necrosis factor alpha (TNF-α). (c.) Blood urea nitrogen (BUN). (d.) Creatinine (CREA). (e.) Total bilirubin (TBIL). (f.) Bilirubin (DBIL). (g.) Alanine aminotransferase (ALT). (h.) Aspartate aminotransferase (AST).

Fig. 4

Continuous photoacoustic in vivo imaging monitoring and analysis within 60 minutes after EB injection at 6 hours post-CLP modeling (n = 4). (a.) Experimental procedures including modeling, EB carrier solution injection, and photoacoustic (PA) imaging. (b.) Sequential PAM images of Sham group post-EB injection. (c.) Sequential PA imaging of CLP group post-EB injection. (d.) Trend of average PA amplitude changes in CLP and Sham groups post-EB injection, EB leakage trends in CLP and Sham groups and EB leakage rates in CLP and Sham groups. (e.) Static comparison of average PA intensity and EB leakage between CLP and Sham groups at 60 minutes. (f.) EB leakage comparison between CLP and Sham groups at 5 minutes. (g.) mean PA intensity comparison between CLP and Sham groups at 60 minutes post-vehicle solution injection. (h.) Photographs of mouse brains at 60 minutes post-EB injection. (i.) Continuous PA imaging after vehicle solution injection at 6 hours post-CLP.

Fig. 5

PA imaging and analysis of vascular morphology in CLP mice at 24 hours and 7 days post-CLP (n = 4). (a.) Representative PA images of cerebral cortex in CLP and sham mice at 24 hours post-CLP. (b.) Representative PA images of cerebral cortex in CLP and sham mice at 7 days post-CLP. (c.) Quantitative analysis of vascular morphology in CLP and sham mice at 24 hours post-CLP. (d.) Quantitative analysis of vascular morphology in CLP and sham mice at 7 days post-CLP.

Fig. 6

Quantitative analysis of blood-brain barrier (BBB) permeability and pathological changes in microvascular morphology across different brain regions. (a.) Schematic division of the Allen Brain Atlas and major brain regions on representative PA images demonstrating EB diffusion. (b.) Quantitative analysis of EB diffusion in different brain regions 6 hours post-CLP. (c.) Quantitative analysis of mean PA intensity in different brain regions 6 hours post-CLP. (d.) Quantitative analysis of vascular morphological changes in primary motor cortex (M1)/secondary motor cortex (M2) brain regions 7 days post-CLP. (e.) Quantitative analysis of vascular morphology in primary visual cortex (V1)/secondary visual cortex (V2) brain regions 7 days post-CLP. (f.) Quantitative analysis of vascular morphology in retrosplenial dysgranular cortex (RSD) brain regions 7 days post-CLP. (g.) Representative PA images showing segmented microvasculature in different brain regions at 7 days post-CLP.

Fig. 7

Motor and cognitive memory performance of Sham and CLP mice (n = 8). (a.) Movement trajectories of Sham and CLP mice in open field test. (b.) Comparison of three behavioral parameters in open field test: total distance, speed, and percentage of time spent in center area (Time in center % of total time). (c.) Comparison of two cognitive parameters in new object recognition (NOR): discrimination index (DI) and recognition index (RI). (d.) Exploration heatmaps of Sham and CLP groups in NOR test.