A total of 20 male (excluding the effects of estrogenic hormones) Sprague-Dawley rats (160–190 g) were used with the permission of the local animal experimental ethics committee. Experimental protocols were conducted in accordance with the guidelines for laboratory animal care and use following the ARRIVE guideline. The rats were freely fed commercial rat food and water sterilized by ⁶⁰Co. Controlled conditions were maintained at 18–22°C, 40–60% relative humidity, and a noise level of < 60 dB at a 12 h (h)/12 h day/night cycle. All subjects were randomized into two groups: sham group (n = 8) or BDL group (n = 12). The sample size was determined by considering the mortality rate according to previous studies (Kong et al., 2016; Luo et al., 2018). Details regarding the BDL procedure are described in the Supplementary Materials. Sham group rats were handled similarly except for the bile duct ligation and abscission (Rodrigo et al., 2010; Kong et al., 2016).

The behavioral studies involved the rotarod test, beam walking test, and motor activity experiment performed 2–3 days prior to the micro-PET scans. The tests were performed to evaluate motor coordination, tolerance, locomotor activity, and vertical activity of the rats (Jover et al., 2006; Rodrigo et al., 2010; Agusti et al., 2011; Kong et al., 2016). Detailed procedures are described in the Supplementary Materials.

Both [¹⁸F]PBR146 and [¹⁸F]DPA-714 can be applied as PET radioligands for TSPO imaging. [¹⁸F]PBR146 and [¹⁸F]DPA-714 were synthesized as previously described (Fookes et al., 2008; Lavisse et al., 2015; Kong et al., 2016; Luo et al., 2018). Procedures for the synthesis of [¹⁸F]DPA-714 and [¹⁸F]PBR146 are provided in the Supplementary Materials. BDL and sham rats were housed in the same conditions for more than 28 days (Butterworth et al., 2009). Micro-PET/CT scans were implemented using an Inveon small animal scanner (Siemens Preclinical Solution) after completion of the behavior studies.

Rats were fixed in the prone position after anesthetization by isoflurane inhalation (induction: 3%, thereafter: 2–2.5%) in oxygen. Radioactive tracers were injected intravenously into the lateral tail vein. PET data were acquired first by CT data acquisition. PET acquisitions were obtained by scanning for 10 min at 50 min post-radiotracer injection, followed by CT scanning for ∼6–10 min to allow for coregistration of radiotracer uptake with tissues. PET setting parameters were as follows: slice thickness = 0.78 mm, matrix size = 128 × 128, field of view (FOV) = 4 cm × 4 cm, and energy levels of acquisition = ∼350–650 keV (default) (Kong et al., 2016). CT setting parameters were as follows: current = 500 A, voltage = 50 kV, and exposure time = 500 ms (Cochran et al., 2017). Micro-PET/CT scans with [¹⁸F]PBR146 and [¹⁸F]DPA-714 were performed for two consecutive days.

Image reconstruction was performed using the Inveon Research Workplace (IRW 3.0, Siemens). PET and CT images were coregistered for correct alignment in three dimensions. Global brain and some organs (e.g., lungs, heart, liver, and kidneys) were drawn on the images for the region of interest (ROI) for quantification. The quantified radioactivity uptake of each ROI was presented as the percent injected dose per gram (%ID/g), calculated by dividing tissue radioactivity with injected dose (assuming the tissue density is 1 g/mL). In addition, regional brain uptake of the radiotracer was evaluated using PMOD software (version 3.7, PMOD Technologies LTD, Zurich, Switzerland). PET images were manually fused with the T2-MRI template after coregistration with CT images. The software then drew 58 ROIs of the brain on the PET images with reference to the MR imaging-based atlas, which yielded corresponding radioactivity concentration values. This processing avoided the effects of peripheral vessels and tissues, which show higher tracer distribution (Cochran et al., 2017; Jiang et al., 2019).

Rats were anesthetized to collect canthus blood (1–2 mL) into procoagulant tubes in the morning, one day prior to micro-PET/CT scans. Blood samples were centrifugalized at 3,000 × g for 10 min (min) at 4°C, and the supernatant was collected and transferred to the local hospital within 2 h for venous blood ammonia measurement. The day after micro-PET/CT scans, blood samples from the heart cavity (5–10 mL) were collected from each rat after deep anesthetization by 10% chloral hydrate. Determination of liver and renal function indicators were conducted in a clinical laboratory. These biochemical indicators from plasma included total bilirubin, direct bilirubin, indirect bilirubin, total protein, albumin, globulin, alanine aminotransferase (ALT), aspartate aminotransferase (AST), alkaline phosphatase, total cholesterol, high-density lipoprotein cholesterol (HDL-C), low-density lipoprotein cholesterol (LDL-C), urea, creatinine, and uric acid (Nanjing Jiancheng Bioengineering Institute, Jiangsu, China). Finally, 5-hydroxytryptamine (5-HT), interferon-γ (IFN-γ), interleukin 1β (IL-1β), IL-6, IL-10, and tumor necrosis factor alpha (TNF-α) were analyzed from plasma samples using enzyme-linked immune sorbent assay (ELISA) kits (Nanjing Jiancheng Bioengineering Institute, Jiangsu, China).

After blood samples were collected, all rats were transcardially perfused with 100–150 mL of saline, followed by 150 mL of phosphate buffered saline (PBS, pH = 7.4) for 20–30 min. The liver, brain, jejunum, ileum, and colon specimens were removed, fixed in 10% buffered formaldehyde, paraffin-embedded, and sliced. The specimens were stained with hematoxylin-eosin (H&E) as previously described (Kong et al., 2016; Luo et al., 2018). In addition, microglia were stained by performing immunohistochemistry of glial fibrillary acidic protein (GFAP, 1:800) and ionized calcium binding adaptor molecule-1 (IBA-1, 1:500) (Wuhan servicebio technology CO.,LTD, Hubei, China) (Dickens et al., 2014; Israel et al., 2016). All histopathological and immunohistochemical slices were scanned using a pathological section scanner. Morphological analysis and cell counts were performed using CaseViewer 1.4 and ImageJ software.

Data analysis was performed using SPSS 20.0 statistical software (SPSS Inc, Chicago, IL, United States). The normality of continuous variables was confirmed by performing a Shapiro-Wilk test. The normally distributed quantitative variables were expressed as mean ± standard deviation (SD), and the median and interquartile range (IQR) were calculated for non-normally distributed data. Independent-sample t-tests and Mann-Whitney tests were performed for sham and BDL group comparisons when appropriate (Kong et al., 2016; Luo et al., 2018). The relationship between behavioral data and radiotracer uptake values in the rat brains was investigated using Spearman’s rank correlation test. P-values of <0.05 were deemed statistically significant.

All eight rats in the sham group were alive during the whole experiment, while four rats in the BDL group died due to a biliary fistula or an anesthesia accident. Weights of the sham group rats were significantly higher than the BDL group rats from the first day after operation to the end of the experiment (Supplementary Figure 1). Results of the rotarod test, beam walking test, and motor activity were significantly different between the sham and BDL groups (P < 0.05, Table 1), indicating that BDL rats developed significant impairments in coordination, tolerance, and motor activity. Serum ammonia levels in the BDL group (39.43 ± 7.91 μmol/L) were higher than those in the sham group (29.88 ± 4.22 μmol/L, P = 0.026). Plasma bilirubin, albumin, globulin, ALT, and AST levels in the BDL group were significantly higher than those in the sham group (P < 0.01, Table 1).

There were significant differences between the two groups regarding plasma total cholesterol, HDL-C, and LDL-C levels (P < 0.01, Table 1). Plasma urea and uric acid levels showed significant differences between the sham and BDL groups (P = 0.032 and P = 0.002, respectively), while plasma creatinine levels showed no significant differences between groups (P = 0.442, Table 1). Plasma 5-HT, IFN-γ, IL-1β, IL-6, IL-10, and TNF-α levels in the BDL group were significantly higher than those in the sham group (P < 0.01, Table 1). These biochemical measurements indicate liver impairments in the rats who underwent BDL.

[¹⁸F]DPA-714 and [¹⁸F]PBR146 micro-PET/CT imaging was conducted on two consecutive days. Although the mean injected radioactivity of [¹⁸F]DPA-714 (279.3 ± 16.7 μci) and [¹⁸F]PBR146 (245.9 ± 19.2 μci) showed significant differences (P < 0.001), the injected radioactivities of [¹⁸F]DPA-714 and [¹⁸F]PBR146 in the sham group (270.9 ± 13.9 and 235.7 ± 19.4 μci, respectively) were not significantly different from the BDL group (286.5 ± 16.4 and 256.1 ± 13.4 μci; P = 0.094 and P = 0.060, respectively). All six rats in the sham group were subject to [¹⁸F]DPA-714 and [¹⁸F]PBR146 micro-PET/CT imaging. Seven rats in the BDL group were subject to [¹⁸F]DPA-714 micro-PET/CT imaging, while [¹⁸F]PBR146 micro-PET/CT scanning was conducted successfully on six BDL group rats, except that one rat died due to an anesthesia accident.

The average%ID/g values in the global brain of [¹⁸F]DPA-714 and [¹⁸F]PBR146 in the BDL rats were higher than those in the sham rats (P < 0.001, Figures 1A,B and Supplementary Table 1). There were significant differences in [¹⁸F]DPA-714 and [¹⁸F]PBR146 uptake values of the lung, liver, and kidney between the BDL and sham groups (P < 0.05), while there were no significant differences in the myocardium (P > 0.05, Figures 1A,B and Supplementary Table 1). Comparison results grouped by [¹⁸F]DPA-714 and [¹⁸F]PBR146 are shown in Table 2. Most organs showed no significant differences (P > 0.05), except that the livers of sham rats showed statistical differences between the two groups (P < 0.001, Table 2).

The comparison of [¹⁸F]DPA-714 and [¹⁸F]PBR146 uptake values in regional brain areas showed consistent results (Supplementary Tables 2, 3). Significant differences among brain regions between the sham and BDL groups were mainly located in the basal ganglia, cingulate cortex, auditory cortex, motor cortex, somatosensory cortex, hippocampus, thalamus, midbrain, and medulla (P < 0.05, Figure 2). The average%ID/g values of the BDL group were higher than the sham group in the brain regions without statistical differences (P > 0.05, Supplementary Tables 2, 3).

Liver weights in the sham group (17.87 ± 1.76 g) were significantly lower than the BDL group (24.71 ± 6.44 g, P = 0.048), suggesting that the livers of rats in the BDL group showed compensatory hyperplasia. Liver H&E staining in the sham group showed normal hepatic structure, while staining in the BDL group revealed biliary cirrhosis with destroyed hepatic cords, expanded bile ducts, and inflammatory infiltration. Additionally, the jejunum, ileum, and colon tissues showed normal intestinal structures in the sham and BDL groups (Figure 3). Brain tissue after H&E staining showed no significant differences between the sham group and BDL group, while histopathological changes were observed during GFAP and IBA-1 immunohistochemical analysis of brain microglia. In particular, microglial cells in the sham rats showed ramified shapes (resting microglia), while they showed amoeboid shapes (activated microglia) in the BDL rats (Figure 3). The sham group showed significantly lower amounts of IBA-1 immune reactive microglia in the basal ganglia than BDL rats (P = 0.006, Table 1).

Duration of time spent on the rotarod and number of crossovers during the motor activity test showed no correlation with [¹⁸F]DPA-714 uptake values (P > 0.05), while the beam crossing time showed a positive correlation with [¹⁸F]DPA-714 uptake values in the global brain (r = 0.571, P = 0.042), including structures such as the bilateral accumbens, striatum, auditory cortex, cingulate cortex, insular cortex, medial prefrontal cortex, orbitofrontal cortex, visual cortex, olfactory, right amygdala, motor cortex, retrosplenial cortex, hippocampus anterodorsal, hippocampus posterior, hypothalamus, and colliculus superior (P < 0.05, Supplementary Table 4).

Beam walking cross time showed no correlation with [¹⁸F]PBR146 uptake values, and the behavioral results showed no correlation with [¹⁸F]PBR146 uptake values in the global brain (P > 0.05, Supplementary Table 5). Duration of time spent on the rotarod showed a negative correlation with [¹⁸F]PBR146 uptake values in the bilateral auditory cortex, cingulate cortex, entorhinal cortex, motor cortex, hypothalamus, left accumbens, medial prefrontal cortex, retrosplenial cortex, hippocampus antero dorsal, right amygdala, frontal association cortex, somatosensory cortex, hippocampus posterior, and olfactory (P < 0.05, Supplementary Table 5). The number of crossovers during the motor activity test showed a negative correlation with [¹⁸F]PBR146 uptake values in the bilateral entorhinal cortex, insular cortex, olfactory, left amygdala, hypothalamus, and right accumbens (P < 0.05, Supplementary Table 5).