Abstract
Purpose
To evaluate the monitoring value of T1 and T2 mapping in assessing kidney injury associated with chronic liver disease and the therapeutic efficacy of bone marrow mesenchymal stem cells (BMSCs) treatment.
Methods
Thirty-six rats were divided into 6 subgroups (n = 6/group) and underwent MRI scanning at 0, 2, 4, 6, 8, and 12 weeks, respectively, followed by biochemical and histological analyses. Seven rats underwent continuous MRI scanning to monitor changes in imaging parameters. Twenty-four rats divided into BMSC and control group. Six rats per group were subjected to serial MRI scans at weeks 13, 14, 15, and 16, another six rats per group were scanned at week 14 and then sacrificed for biochemical and renal histological analysis.
Results
From baseline to 12 weeks, renal hematoxylin and eosin (HE) scores and α-smooth muscle actin (α-SMA) levels increased significantly, and similar trends were found in renal T1 and T2 values. Following BMSCs injection, both BMSC and control groups exhibited reductions in HE scores and α-SMA levels, with BMSC group demonstrating more substantial decreases. Renal T1 and T2 values declined in both groups, with the BMSC group showing significantly lower T2 values than the control group. Strong correlations were found between renal T1/T2 values and HE scores, α-SMA levels (|r| = 0.419–0.724). The area under the curve values for T1 and T2 in differentiating renal injury severity across different renal strips were from 0.793 to 0.930.
Conclusion
T1 and T2 mapping can effectively monitor renal injury progression in chronic liver disease, with T2 values demonstrating greater potential for assessing the therapeutic efficacy of BMSCs.
Keywords: chronic liver disease, magnetic resonance imaging, renal injury, T1 mapping, T2 mapping
Introduction
Chronic liver disease affects a substantial global patient population, often leading to poor prognosis due to frequent late-stage complications. Clinical evidence demonstrates that liver cirrhosis patients develop hemodynamic disturbances, including portal hypertension, which causes venous congestion and reduced systemic blood volume. This impaired renal perfusion elevates the risk of both acute kidney injury and chronic kidney disease.1 Furthermore, existing treatment options for established renal damage remain limited and show minimal impact on patient outcomes. Consequently, careful monitoring of renal burden in chronic liver disease patients is essential.
Due to the absence of renal pathological testing, it remains unclear whether the kidneys remain intact during the stage of liver fibrosis or if they have already become more vulnerable. Furthermore, patients with chronic liver disease frequently exhibit sarcopenia, which can lead to inaccuracies in renal function estimation based on creatinine levels, potentially resulting in an overestimation of renal function.2 In recent years, novel biomarkers such as cystatin C, kidney injury molecule-1, and neutrophil gelatinase–associated lipocalin have emerged as effective indicators of renal injury; however, their specificity remains relatively low, and they have not yet been widely adopted in clinical practice.3 T1 and T2 mapping are quantitative MRI techniques that enable for the assessment of tissue characteristics by measuring variations in T1 and T2 relaxation times.4 Recent studies have demonstrated that T1 and T2 mapping can evaluate renal injury in various models. Relevant researches5–10 have demonstrated that T2 mapping can effectively assess liver fibrosis, myocardial injury, and renal tissue edema. Additionally, one study11 indicated that T2 mapping can accurately identify renal damage in deoxycorticosterone acetate–salt hypertensive rats. Research by Hueper et al.12 demonstrated that T1 mapping provides a noninvasive method to assess the severity of ischemia-reperfusion–related acute kidney injury in mice and predict subsequent outcomes. However, the diagnostic value of T1 and T2 mapping in chronic liver disease–related kidney injury and their utility in evaluating the efficacy of stem cell therapy remains to be investigated.
Stem cell therapy has been widely applied in renal diseases and has demonstrated promising efficacy. Several studies13–15 have shown that mesenchymal stem cells can mitigate renal oxidative damage, apoptosis, and fibrosis by enhancing renal perfusion and capillary density. However, further exploration is needed to optimize methods for assessing the efficacy of stem cell treatment. Consequently, developing noninvasive diagnostic tests for early diagnosis, evaluation, and monitoring of renal pathology, as well as predicting renal function prognosis, holds significant importance for improving treatment outcomes and long-term prognosis in patients with chronic liver disease–related kidney injury.
Therefore, our objective was to investigate longitudinal changes in renal injury during the progression of chronic liver disease using T1 and T2 mapping and to evaluate alterations in renal injury following bone marrow mesenchymal stem cell (BMSC) therapy.
Materials and Methods
All experiments were approved by the Animal Experimentation Committee of Nankai University (2022-SYDWLL-000236) and conducted in accordance with the Guide for the Care and Use of Laboratory Animals.
Experimental animals
A total of 67 healthy male Sprague-Dawley (SD) rats (aged 6–8 weeks) were included in this study. The rats were obtained from Beijing Huafukang Biotechnology and housed in the Key Laboratory of Transplantation Research at Nankai University. Following a 1-week acclimatization period, 36 rats were randomly selected to receive subcutaneous injections of carbon tetrachloride (CCl4) in the nape of the neck to establish a chronic liver disease model (CCl4: corn oil ratio of 2:3, administered twice weekly with an initial dose of 5 mL/kg and a maintenance dose of 3 mL/kg). These rats were divided into 6 subgroups based on modeling time points (0, 2, 4, 6, 8, and 12 weeks), with 6 rats per group. After renal T1 and T2 mapping scans, the rats were euthanized for venous blood collection for biochemical analysis, and the right kidney was harvested for histological examination. Additionally, 7 rats were randomly selected for chronic liver disease model establishment and underwent continuous dynamic T1 and T2 mapping scans at 0, 2, 4, 6, 8, and 12 weeks to assess renal T1 and T2 mapping changes (Fig. 1).
Fig. 1.
Experimental design diagram.BMSC, boe marrow mesenchymal stem cell; CCl4, carbon tetrachloride; PBS, phosphate-buffered saline.
Furthermore, 24 rats were subcutaneously injected with CCl4 for 12 weeks to establish a chronic liver disease model (using the same method as previously described). These rats were randomly divided into 2 groups: the BMSC group (n = 12) and the control group (n = 12). At week 12, rats in the BMSC group received an injection of 3 × 106 BMSCs. Six rats underwent continuous dynamic scanning at weeks 12, 13, 14, 15, and 16 before euthanasia for biochemical and renal histological analysis, while the remaining 6 rats were euthanized after MRI scanning at week 14 for blood biochemical and renal histological analysis. Rats in the control group received an equal volume of phosphate-buffered saline (PBS) at week 12, with all other procedures identical to those in the BMSC group (Fig. 1).
Preparation of BMSCs
The BMSCs were prepared according to our previously published method.16 Briefly, BMSCs were isolated from male SD rats weighing 100g–120g. Bone marrow was flushed from the femurs using a solution containing 10% fetal bovine serum and 1% penicillin/streptomycin. The cells were resuspended in complete culture medium and incubated at 37°C in a humidified atmosphere with 5% CO2. The medium was replaced every 72 hour. Upon reaching 80%–90% confluence, third-generation cells were trypsinized, washed, centrifuged, and resuspended in PBS at a concentration of 1 × 107 cells/mL. Following adipogenic and osteogenic differentiation induction, the BMSCs exhibited robust differentiation capacity.
MRI examination
MRI scans were performed using a 3T MRI scanner (MAGNETOM Prisma; Siemens Healthineers, Forchheim, Germany) equipped with a 16-channel small animal coil (Chenguang, Shanghai, China). Prior to scanning, the rats were fasted for 8 hour to minimize gastrointestinal interference. T2-weighted (T2W), T1 mapping and T2 mapping images were acquired, the parameters are as follows: T2-weighted images (T2WI) with the following parameters: TR = 5400 ms, TE = 99 ms, flip angle = 148 deg, fat suppression = fat saturation, FOV = 80 × 80 mm2, slice thickness = 1.5 mm, matrix = 192 × 192, voxel size = 0.42 × 0.42 × 1.50 mm3, bandwidth = 250 Hz/px, and acquisition time = 3 mins 05s. A B1 mapping sequence was incorporated prior to T1 mapping to correct for magnetic field inhomogeneity, then T1 mapping was obtained using a 3D gradient echo sequence with the Volumetric Interpolated Breath-hold Examination (VIBE) technique and a dual-flip-angle method (TR = 6.79 ms, TE = 2.85 ms, flip angles = 2.0 deg and 10 deg, FOV = 80 × 60 mm2, slice thickness = 2.0 mm, matrix = 256 × 230, voxel size = 0.31 × 0.26 × 2.00 mm3, bandwidth = 260 Hz/px, acquisition time = 4 mins 14s). T2 mapping was performed using a multi-echo spin echo sequence (TR = 2000 ms, TE = 13/26/39/52/65/78 ms, FOV = 120 × 120 mm2, slice thickness = 3 mm, matrix = 128 × 96, reconstructed voxel size = 0.94 × 1.25 × 3.00 mm3, bandwidth = 201 Hz/px, acquisition time = 3 mins 38s). During imaging, the rats were positioned supine to reduce respiratory motion artifacts. Anesthesia was maintained throughout the procedure using a small animal gas anesthesia system (Yuyan Scientific Instrument, Shanghai, China).
Image data processing and analysis
After image acquisition, the images were directly transmitted to a Siemens post-processing workstation (Siemens Healthcare, Erlangen, Germany). With reference to the axial T2WI structural images of the kidneys (Fig. 2a), the kidney was divided into 3 zones: the cortex (CO), outer stripe of the outer medulla (OSOM), and inner stripe of the outer medulla (ISOM). On the T1 mapping and T2 mapping images, a circular ROI tool was manually used to delineate ROIs in the renal CO, OSOM, and ISOM, respectively. The principles for ROI delineation were as follows: ①The largest cross-sectional slice of the kidney was selected; ②Areas with image artifacts were avoided; ③Five ROIs were delineated in each zone; ④Efforts were made to keep the positions of ROIs consistent between the T1 mapping and T2 mapping images for each corresponding zone; ⑤The size of each ROI was maintained at approximately 4 mm2. The mean value of the 5 ROIs was taken as the final result for each zone. To evaluate inter-observer agreement, 2 radiologists (Observer 1: 13 years of experience; Observer 2: 6 years of experience) independently delineated the ROIs while blinded to histological findings.
Fig. 2.
(a) Schematic diagram of renal T2WI illustrating the division into the CO, OSOM, and ISOM. (b) Comparison of BUN levels at 0, 2, 4, 6, 8, and 12 weeks. (c) Comparison of BUN levels between the BMSC-treated group and the control group at 12, 14, and 16 weeks. (d) Comparison of Scr levels at 0, 2, 4, 6, 8, and 12 weeks. (e) Comparison of Scr levels between the BMSC-treated group and the control group at 12, 14, and 16 weeks.BMSC, bone marrow mesenchymal stem cell; BUN, blood urea nitrogen; CO, cortex; ISOM, inner stripe of the outer medulla; OSOM, outer stripe of the outer medulla; Scr, serum creatinine; T2WI, T2-weighted imaging.
Biochemical analysis
After completing the MRI scans at the designated time points, 3 mL of venous blood was collected from the inferior vena cava of the anesthetized rats. The blood samples were centrifuged at 3000 rpm for 10 mins at 4°C to obtain serum. Serum levels of creatinine (Scr), blood urea nitrogen (BUN), alanine aminotransferase (ALT), and aspartate aminotransferase (AST) were measured using a URIT-8210 fully automated biochemical analyzer (Yuqing Technology Medical Electronics, Guilin, Guangxi, China).
Histological analysis
After completing MRI scans at the designated time points, the right kidney tissues of anesthetized rats were harvested. The kidneys were fixed in 10% neutral buffered formalin for 24 hour, routinely processed, paraffin-embedded, and sectioned into 2 μm-thick slices for staining. Hematoxylin and eosin (HE) staining and α-smooth muscle actin (α-SMA) immunohistochemical staining were performed on the kidney tissues. Kidney injury was scored according to the criteria established by Wang et al.4 Under a Nikon microscope (Nikon, Tokyo, Japan), 5 FOVs were selected in each of the CO, OSOM, and ISOM regions for HE scoring based on the area of injury (0, normal; 1, ≤ 5%; 2, 5%–25%; 3, 25%–50%; 4, 50%–75%; 5, 75%–100%). The mean score of the 5 fields was taken as the final HE score for each zone. The scoring criteria included cellular swelling, vacuolization, brush border loss, tubular casts, tubular dilation, interstitial vascular dilation and congestion, interstitial fibrosis, and tubular necrosis. α-SMA expression was semi-quantitatively analyzed using ImageJ software (National Institutes of Health, Bethesda, MD, USA). Two blinded clinical pathologists (Zhandong Hu and Dan Tong, with 12 and 10 years of renal pathology experience, respectively) independently performed the evaluations without knowledge of each other’s assessments or the experimental design. If discrepancies arose between the evaluations of the 2 experts, the final result was determined after discussion.
Statistical analysis
Statistical analyses and graphical representations were performed using GraphPad Prism (version 9.0; GraphPad, La Jolla, CA, USA) and Hiplot Pro (https://hiplot.com.cn/home/index.html). All measured data are presented as mean ± standard deviation. Intraclass correlation coefficients (ICCs) were employed to evaluate the reliability of T1 and T2 measurements. Dynamic imaging data (T1 and T2 values) were evaluated using repeated-measures analysis of variance (ANOVA), with post-hoc multiple comparisons performed using the Bonferroni method. Biochemical and pathological data at continuous time points were analyzed using one-way ANOVA. Two-way ANOVA was used to compare imaging, biochemical, and pathological parameters among groups at each time point. Pearson correlation analysis was applied for normally distributed data, whereas Spearman correlation analysis was used for non-normally distributed data. Receiver operating characteristic curve analysis was performed using MedCalc (version 19.5.3; MedCalc Software, Ostend, Belgium) to evaluate the diagnostic efficacy for kidney injury. A P-value of < 0.05 was considered statistically significant.
Results
Animal models
Following the initiation of model construction, liver fibrosis progressively developed, with pseudolobule formation and marked liver cirrhosis observed in the advanced stages.
Liver and kidney biochemical indicators
In the first part, ALT levels were consistently elevated from weeks 2 to 12 compared to week 0 (P < 0.05), while AST levels showed significant increases at weeks 4, 8, and 12 (P < 0.01) (Supplemental Figure 1). No significant changes in BUN were observed from weeks 0 to 12 (P > 0.05) (Fig. 2b). In the second part, both the BMSC and control groups demonstrated decreased ALT levels at weeks 14 and 16 compared to week 12 (P < 0.0001), with no significant differences between groups (P > 0.05) (Supplemental Fig. 1). Similarly, AST levels declined in both groups at weeks 14 and 16 (P < 0.0001) (Supplemental Fig. 1). BUN levels in the BMSC group were lower at weeks 14 and 16 than at week 12 (P < 0.01), whereas no significant changes occurred in the control group (P > 0.05) (Fig. 2c). In the first part, Scr levels exhibited a gradual increase as the model progressed, with only the 8 week time point showing a statistically significant difference from baseline (P < 0.05) (Fig. 2d). In the second part, Scr levels in the BMSC group were lower at weeks 14 and 16 than at week 12 (P < 0.05). In the control group, Scr decreased only at week 14 and slightly increased by week 16, though no significant differences were observed between groups at any time point (P > 0.05) (Fig. 2e).
Renal histology
In the first part, HE scores of the kidney across various regions from weeks 2 to 12 were significantly higher than baseline (P < 0.0001). The HE scores progressively increased from weeks 0 to 4, exhibited a slight decline at week 6, and then rose markedly again from weeks 8 to 12. Concurrently, α-SMA levels in different kidney regions gradually increased from weeks 0 to 12 (P < 0.05) (Table 1, Fig. 3a–d). In the second part, both the BMSC and control groups demonstrated reduced HE scores compared to week 12 (P < 0.0001), with the BMSC group showing a more significant reduction than the control group (P < 0.05). The BMSC group exhibited a substantial decrease in α-SMA levels at weeks 14 and 16 compared to week 12 (P < 0.0001), whereas the control group showed a mild reduction at week 16 (P < 0.01) and an increase at week 14 relative to week 12 (Table 2, Fig. 3e–h).
Table 1.
Changes in kidney HE scores and α-SMA within 12 weeks of carbon tetrachloride injection
| 0 W | 2 W | 4 W | 6 W | 8 W | 12 W | ||
|---|---|---|---|---|---|---|---|
| HE score | CO | 0.00 ± 0.00 | 0.33 ± 0.21**** | 3.20 ± 0.47**** | 2.87 ± 0.21**** | 2.97 ± 0.27**** | 3.97 ± 0.20**** |
| OSOM | 0.00 ± 0.00 | 0.93 ± 0.10**** | 1.57 ± 0.15**** | 1.37 ± 0.15**** | 2.47 ± 0.21**** | 3.27 ± 0.10**** | |
| ISOM | 0.00 ± 0.00 | 0.60 ± 0.13**** | 1.33 ± 0.16**** | 1.17 ± 0.15**** | 1.67 ± 0.16**** | 2.30 ± 0.30**** | |
| α-SMA | CO | 0.02 ± 0.00 | 0.28 ± 0.07**** | 0.85 ± 0.07**** | 1.14 ± 0.08**** | 1.28 ± 0.04**** | 2.40 ± 0.13**** |
| OSOM | 0.04 ± 0.01 | 0.36 ± 0.06* | 1.10 ± 0.16**** | 1.26 ± 0.03**** | 1.59 ± 0.19**** | 3.14 ± 0.36**** | |
| ISOM | 0.07 ± 0.01 | 0.42 ± 0.06* | 1.44 ± 0.05**** | 1.42 ± 0.06**** | 2.23 ± 0.08**** | 3.80 ± 0.50**** |
*P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001 compared with 0 W.
CO, cortex; HE, hematoxylin-eosin staining; ISOM, inner stripe of the outer medulla; OSOM, outer stripe of the outer medulla; α-SMA, alpha-smooth muscle actin.
Fig. 3.
(a) Comparison of renal HE scores (200× magnification) at different time points (0–12 weeks). (b) Comparison of renal α-SMA expression levels (200× magnification) at different time points (0–12 weeks). (c) Comparison of HE scores in the renal cortex between weeks 0 and 12. (d) Comparison of α-SMA expression levels in the renal cortex between weeks 0 and 12. (e) Comparison of changes in HE scores in the renal cortex between the BMSC group and the control group. (f) Comparison of changes in α-SMA expression levels in the renal cortex between the BMSC group and the control group. (g) Comparison of renal HE scores (200× magnification) between the BMSC group and the control group at weeks 14 and 16. (h) Comparison of renal α-SMA expression levels (200× magnification) between the BMSC group and the control group at weeks 14 and 16.BMSC, bone marrow mesenchymal stem cell; HE, hematoxylin and eosin; α-SMA, α-smooth muscle actin.
Table 2.
HE scores and α-SMA changes in the kidney at 12, 14, and 16 weeks for the BMSC group and control group
| 12 W | 14 W | 16 W | |||
|---|---|---|---|---|---|
| HE score | CO | BMSC Group | 3.97 ± 0.20 | 2.40 ± 0.13$$$$**** | 2.03 ± 0.15$$$$** |
| Control Group | 3.97 ± 0.20 | 2.90 ± 0.11#### | 2.33 ± 0.16#### | ||
| OSOM | BMSC Group | 3.27 ± 0.10 | 1.57 ± 0.15$$$$**** | 1.37 ± 0.08$$$$**** | |
| Control Group | 3.27 ± 0.10 | 2.10 ± 0.17#### | 1.83 ± 0.15#### | ||
| ISOM | BMSC Group | 2.30 ± 0.30 | 1.33 ± 0.10$$$$* | 1.17 ± 0.15$$$$** | |
| Control Group | 2.30 ± 0.30 | 1.67 ± 0.16#### | 1.57 ± 0.08#### | ||
| α-SMA | CO | BMSC Group | 2.40 ± 0.13 | 1.22 ± 0.09$$$$**** | 0.57 ± 0.04$$$$**** |
| Control Group | 2.40 ± 0.13 | 2.50 ± 0.13 | 2.20 ± 0.01## | ||
| OSOM | BMSC Group | 3.14 ± 0.36 | 1.45 ± 0.02$$$$**** | 0.67 ± 0.05$$$$**** | |
| Control Group | 3.14 ± 0.36 | 3.35 ± 0.06 | 2.68 ± 0.12## | ||
| ISOM | BMSC Group | 3.80 ± 0.50 | 2.34 ± 0.08$$$$**** | 1.53 ± 0.05$$$$**** | |
| Control Group | 3.80 ± 0.50 | 3.95 ± 0.12 | 2.76 ± 0.09#### |
$P < 0.05, $$P < 0.01, $$$P < 0.001, $$$$P < 0.0001 compared with 12w.
#P < 0.05, ##P < 0.01, ###P < 0.001, ####P < 0.0001 compared with 12w.
*P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001 compared with Control Group.
BMSC, bone marrow mesenchymal stem cell; CO, cortex; HE, hematoxylin-eosin staining; ISOM, inner stripe of the outer medulla; OSOM, outer stripe of the outer medulla; α-SMA, alpha-smooth muscle actin.
Characteristics of T1 and T2 value changes
For validation of T1/T2 scanning stability and reproducibility, healthy rats underwent 2 renal scans at a 1 week interval. ICCs > 0.8 between the 2 scan sets (Supplementary Tables 1–2) confirmed good technical stability and strong reproducibility.
In the first phase of the experiment, T1 values across different kidney regions demonstrated a progressive increase from weeks 0 to 12 (Table 3, Fig. 4a, b). Specifically, T1 values in the renal CO and OSOM showed a significant rise from weeks 0 to 12 (P < 0.01). In the ISOM, T1 values increased from weeks 0 to 4 (P < 0.05), exhibited a slight decrease at week 6, and then continued to rise at weeks 8 and 12 (P < 0.01). During the second phase, T1 values in the renal CO of both the BMSC and control groups progressively declined at weeks 14, 15, and 16 (all P < 0.05). In the OSOM and ISOM, the BMSC group exhibited lower T1 values at weeks 13, 14, 15, and 16 compared to week 12 (all P < 0.05). In the control group, T1 values in the OSOM were lower at weeks 14 and 16 (all P < 0.05), whereas in the OSOM, only the T1 value at week 16 was significantly reduced (P < 0.05). A statistical difference in the T1 values between the 2 groups was observed solely in the ISOM at week 15 (P < 0.05) (Table 4, Fig. 4a, c).
Table 3.
Changes in kidney T1 and T2 values within 12 weeks of carbon tetrachloride injection
| 0 W | 2 W | 4 W | 6 W | 8 W | 12 W | ||
|---|---|---|---|---|---|---|---|
| T1-value (ms) | CO | 1378.97 ± 15.61 | 1472.25 ± 41.02** | 1548.11 ± 31.58*** | 1552.85 ± 76.18** | 1583.83 ± 50.42*** | 1640.00 ± 56.16*** |
| OSOM | 1324.48 ± 35.70 | 1447.60 ± 28.90** | 1481.30 ± 50.45** | 1522.56 ± 71.40** | 1529.25 ± 51.52** | 1556.18 ± 31.30*** | |
| ISOM | 1465.06 ± 50.23 | 1613.60 ± 71.86* | 1647.57 ± 52.89* | 1623.00 ± 119.85 | 1632.88 ± 66.86** | 1663.02 ± 61.51** | |
| T2-value (ms) | CO | 72.97 ± 3.20 | 86.19 ± 4.72** | 98.97 ± 4.80**** | 99.93 ± 2.50**** | 101.79 ± 3.66**** | 1.6.28 ± 8.09*** |
| OSOM | 66.73 ± 3.21 | 76.23 ± 4.18** | 82.48 ± 3.34**** | 83.81 ± 2.69**** | 86.65 ± 4.63**** | 86.47 ± 3.98*** | |
| ISOM | 75.69 ± 5.27 | 88.31 ± 3.74** | 96.90 ± 5.52*** | 97.80 ± 4.59**** | 103.08 ± 6.50**** | 102.41 ± 3.86*** |
*P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001 compared with 0 W.
CO, cortex; ISOM, inner stripe of the outer medulla; OSOM, outer stripe of the outer medulla.
Fig. 4.
(a–c) Temporal changes in renal T1 values at different time points (0–12 weeks), along with comparisons of T1 values between the BMSC group and control group at each time point. (d–f) Temporal changes in renal T2 values at different time points (0–12 weeks), along with comparisons of T2 values between the BMSC group and control group at each time point.BMSC, bone marrow mesenchymal stem cell.
Table 4.
Changes in kidney T1 and T2 values in the BMSC group and control group between 12 and 16 weeks
| 12 W | 13 W | 14 W | 15 W | 16 W | |||
|---|---|---|---|---|---|---|---|
| T1-value (ms) | CO | BMSC Group | 1618.80 ± 39.64 | 1505.23 ± 34.56 | 1455.36 ± 30.79$$ | 1407.32 ± 37.26$$$$ | 1379.74 ± 36.83$$$$ |
| Control Group | 1609.25 ± 52.56 | 1559.79 ± 93.01 | 1453.95 ± 109.97## | 1484.79 ± 107.55# | 1414.12 ± 76.16### | ||
| OSOM | BMSC Group | 1541.99 ± 43.12 | 1422.15 ± 33.94$$ | 1382.45 ± 24.35$$$$ | 1354.57 ± 14.80$$$$ | 1333.00 ± 13.82$$$$ | |
| Control Group | 1508.42 ± 45.79 | 1457.06 ± 69.02 | 1407.42 ± 39.85# | 1431.71 ± 116.95 | 1385.95 ± 51.67## | ||
| ISOM | BMSC Group | 1669.32 ± 17.97 | 1525.46 ± 64.03$$ | 1466.91 ± 28.16$$$$ | 1451.34 ± 39.45$$$$* | 1469.07 ± 25.09$$$$ | |
| Control Group | 1637.67 ± 40.66 | 1584.27 ± 56.66 | 1537.06 ± 111.72 | 1549.14 ± 91.93 | 1517.61 ± 78.66# | ||
| T2-value (ms) | CO | BMSC Group | 104.54 ± 6.63 | 86.87 ± 1.81$$$$ | 79.57 ± 3.97$$$$* | 74.36 ± 2.24$$$$**** | 68.97 ± 3.03$$$$**** |
| Control Group | 99.48 ± 5.47 | 89.30 ± 7.13## | 86.31 ± 2.92#### | 87.26 ± 2.85### | 84.05 ± 2.33#### | ||
| OSOM | BMSC Group | 85.09 ± 9.44 | 77.44 ± 3.66$ | 73.44 ± 4.83$$$** | 69.82 ± 2.16$$$$**** | 66.75 ± 2.71$$$$**** | |
| Control Group | 84.48 ± 2.84 | 80.94 ± 4.01 | 82.22 ± 2.22 | 83.71 ± 3.16 | 81.60 ± 2.69 | ||
| ISOM | BMSC Group | 99.79 ± 2.01 | 91.31 ± 2.27$$ | 86.95 ± 6.54$$$$* | 84.41 ± 3.33$$$$**** | 81.33 ± 1.52$$$$**** | |
| Control Group | 98.84 ± 3.09 | 96.04 ± 3.90 | 93.22 ± 3.26 | 95.09 ± 4.44 | 92.57 ± 3.09# |
$P < 0.05, $$P < 0.01, $$$P < 0.001, $$$$P < 0.0001 compared with 12w.
#P < 0.05, ##P < 0.01, ###P < 0.001, ####P < 0.0001 compared with 12w.
*P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001 compared with Control Group.
BMSC, bone marrow mesenchymal stem cell; CO, cortex; ISOM, inner stripe of the outer medulla; OSOM, outer stripe of the outer medulla.
In the first part of the experiment, T2 values increased progressively across all kidney regions (all P < 0.01) (Table 3, Fig. 4d, e). In the second part, T2 values in the CO decreased at weeks 13, 14, 15, and 16 compared with week 12 in both the BMSC and control groups (all P < 0.05). In the OSOM, the T2 values declined in the BMSC group at weeks 13, 14, 15, and 16 (all P < 0.05), whereas no significant changes were observed in the control group (all P > 0.05). In the ISOM, T2 values decreased in the BMSC at weeks 13, 14, 15, and 16 (all P < 0.05), while the control group exhibited a reduction only at week 16 (P < 0.05). Additionally, at weeks 14, 15, and 16, T2 values in the BMSC group were significantly lower than those in the control group (all P < 0.05) (Table 4, Fig. 4d, f).
Correlation between T1, T2 values and renal histological parameters
T1 values in the renal CO demonstrated a moderate positive correlation with HE scores (r = 0.7024, P < 0.0001) and a weak correlation with α-SMA levels (r = 0.4116, P = 0.0011). In the OSOM, T1 values showed weak positive correlations with both HE scores (r = 0.5624, P < 0.0001) and α-SMA levels (r = 0.4456, P = 0.0004). T2 values in the renal CO exhibited a strong positive correlation with HE scores (r = 0.7909, P < 0.0001), while T2 values in the renal OSOM and ISOM displayed moderate positive correlations with HE scores (|r| = 0.6258–0.6948, P < 0.0001). Additionally, T2 values in the renal CO and OSOM showed moderate positive correlations with α-SMA levels (|r| = 0.5069–0.6199, P < 0.0001), whereas T2 values in the ISOM demonstrated a weak positive correlation with α-SMA levels (r = 0.4189, P < 0.0009) (Table 5, Fig. 5).
Table 5.
Correlation between renal regional zones and pathological parameters
| T1 CO | T1 OSOM | T1 ISOM | T2 CO | T2 OSOM | T2 ISOM | |||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| r | P | r | P | r | P | r | P | r | P | r | P | |
| HE score | 0.7374 | < 0.0001 | 0.4758 | 0.0001 | 0.3008 | 0.0195 | 0.7796 | < 0.0001 | 0.6413 | < 0.0001 | 0.6687 | < 0.0001 |
| α-SMA | 0.3620 | 0.0045 | 0.3575 | 0.005 | 0.0217 | 0.8696 | 0.4637 | 0.0002 | 0.6237 | < 0.0001 | 0.4723 | 0.0001 |
The P-values were obtained through correlation analysis using the Spearman.
CO, cortex; HE, hematoxylin-eosin staining; ISOM, inner stripe of the outer medulla; OSOM, outer stripe of the outer medulla; α-SMA, alpha-smooth muscle actin.
Fig. 5.
(a–f) Correlation analysis between renal T1/T2 values and HE score. (g–l) Correlation analysis between renal T1/T2 values and α-SMA.HE, hematoxylin and eosin; α-SMA, α-smooth muscle actin.
Analysis of the diagnostic efficacy for kidney injury
Using a kidney HE score of 2 as the cutoff value based on kidney injury scoring criteria, we classified kidney injury into mild and moderate-to-severe categories. The T1 values in the CO, OSOM, and ISOM differentiated between mild and moderate-to-severe kidney injury, with area under the curve (AUC) values of 0.828, 0.876, and 0.795, respectively. Similarly, the T2 values in the CO, OSOM, and ISOM showed AUC values of 0.888, 0.930, and 0.837, respectively, for distinguishing injury severity. Although T2 values demonstrated higher diagnostic performance than T1 values for classifying mild versus moderate-to-severe kidney injury, no statistically significant difference was observed between the 2 (P > 0.05) (Supplemental Table 3, Fig. 6).
Fig. 6.
(a–c) The diagnostic performance of T1/T2 values in differentiating mild from moderate-to-severe renal injury in the renal CO, OSOM, and ISOM.CO, cortex; ISOM, inner stripe of the outer medulla; OSOM, outer stripe of the outer medulla.
Discussion
Previous studies have demonstrated that T1 and T2 mapping offer distinct advantages in evaluating tissue damage, particularly for detecting tissue edema.17,18 In this study, we employed T1 and T2 mapping to assess renal alterations in rats with chronic liver disease and monitored renal changes following BMSC intervention in cirrhotic rats. Our findings suggest that both T1 and T2 mapping techniques show significant potential for longitudinal monitoring of renal injury progression in chronic liver disease. Furthermore, T2 mapping appears to provide greater sensitivity in evaluating the therapeutic efficacy of BMSCs for renal injury associated with chronic liver disease.
T2 mapping, as a noninvasive technique, has substantial advantages in the assessment of renal tissue edema, inflammation, and fibrotic injury. Our normal renal T2 values were consistent with previous study10,11 which use same imaging sequence with us. While, our renal T2 values were higher than those reported by Wang et al.4 This may, in part, be attributable to vendor-specific pulse sequence differences and the unavoidable inclusion of the first echo in our commercial sequence-based T2 calculation. To ensure technical feasibility and stability, reproducibility experiments were performed in normal rats at different time points prior to the main study. In future work, we will continue to optimize sequence parameters to further improve the accuracy and robustness of T2 mapping for renal injury assessment.
During the 0- to 4 week period of CCl4 injection, tubular dilation was observed in the renal zones, along with mild edema and localized vacuolization of renal tubular epithelial cells. We hypothesize that these changes may reflect the body’s stress response to toxic stimuli or secondary renal effects of liver injury, though the precise mechanisms require further investigation. From weeks 4 to 12, as hepatic fibrosis progressed to cirrhosis, renal injury worsened markedly, characterized by pronounced vacuolization of renal tubular epithelial cells, brush border loss, and tubular cast formation. This deterioration likely resulted from reduced renal perfusion and hypoxia secondary to advancing hepatic fibrosis and cirrhosis. Throughout the 0- to 12-week period, α-SMA levels in the renal zones progressively increased, indicating activation of renal fibrosis pathways as chronic liver disease advanced, leading to fibroblast activation and renal fibrosis development. While decompensated cirrhosis is known to cause renal complications, our experimental results suggest an additional consideration: even during the hepatic fibrosis stage, we observed slight renal α-SMA expression, raising the possibility of early renal injury activation at this stage, which merits further investigation. BMSCs have demonstrated renal protective effects through mechanisms including microcirculation preservation, inflammation reduction, oxidative stress mitigation, and ferroptosis inhibiting.14,15,19,20 In our study, BMSC intervention significantly reduced renal HE scores and α-SMA levels compared to the controls, supporting their reparative role in chronic liver disease–related renal injury, potentially through modulation of fibroblast activation, consistent with prior research findings.
Previous studies have demonstrated that T1 and T2 values can characterize tissue composition changes, particularly in water content,21,22 enabling for the quantification of renal pathological alterations. In the initial phase (weeks 0–4), renal T1 and T2 values increased rapidly, correlating with progressive renal injury as evidenced by histopathological findings, including tubular epithelial cell vacuolization, edema, and loss of brush border loss. These changes likely reflect tissue edema and intracellular/extracellular water retention, leading to prolonged T1 and T2 values, consistent with prior research.23,24 From weeks 4 to 12, the rate of T1 and T2 elevation slowed, possibly due to the reduced edema progression. However, persistent renal α-SMA expression—a key marker of fibrosis in renal and hepatic fibrosis25,26—suggested ongoing interstitial water retention,23 sustaining the gradual rise in relaxation times. Thus, last-stage T1 and T2 changes may reflect both residual edema and fibrotic repair activation. Notably, in the OSOM, T1 values exhibited a transient decline at weeks 4 and 6, while T2 value changes were minimal. This variability may stem from the OSOM’s high tubular density and collecting duct sensitivity to water content fluctuations.
In the second phase of the experiment, we observed that following BMSCs intervention, the renal HE scores progressively decreased at weeks 14 and 16, suggesting an improvement in renal injury. Although the control group also exhibited a reduction in renal HE scores, the extent of decline was less pronounced than in the BMSC group. Changes in renal α-SMA levels closely mirrored those of the renal HE scores. Renal T1 values decreased in both groups at weeks 14 and 16, though no statistically significant differences were noted between them. In the contrast, renal T2 values in the BMSC group were significantly lower at weeks 14 and 16 compared to week 12, with statistically significant differences observed between the groups at these time points. These findings indicate that T2 values may more sensitively detect microscopic renal tissue changes when BMSCs are used to ameliorate renal injury in chronic liver disease, particularly in cases of severe renal injury with marked tissue edema and inflammatory infiltration, as previously demonstrated in studies on ischemia-reperfusion renal injury5 and myocardial injury.27 Collectively, our results suggest that T2 values are more sensitive than T1 values in monitoring mesenchymal stem cell–mediated improvements in renal injury in chronic liver disease, as they appear to reflect subtle microscopic changes in renal injury. This aligns with prior research demonstrating the sensitivity of T2 values to intracellular and extracellular edema in tissue injury.24
T1 and T2 mapping show promise as valuable tools for monitoring renal injury associated with chronic liver disease, demonstrating strong correlations with pathological indicators in assessing renal injury progression and evaluating the effects of BMSC interventions. Notably, T2 mapping offers distinct advantages in detecting pathological changes primarily linked to water content variations. Furthermore, the scanning techniques for T1 and T2 mapping continue to improve, producing more stable results and reinforcing their potential as reliable clinical methods for evaluating chronic liver disease–related renal injury.
Limitations
This study has several limitations. First, the sample size was relatively small. Second, complete voxel-level correspondence between MRI images and histological sections was not achieved, which may have introduced a certain degree of bias. Third, the T2 mapping sequence used in this study was vendor-provided, and the first echo was not excluded, which may have influenced the accuracy of T2 estimation. Nevertheless, the stability of this sequence was confirmed through repeated measurements. Finally, all experiments were conducted exclusively at 3T; variations in magnetic field strength may affect quantitative parameter values. Future studies will focus on expanding the sample size, optimizing scanning protocols, and achieving voxel-level imaging–histology matching to improve noninvasive histopathological assessment. In addition, the effects of different static magnetic field strengths on imaging parameters will be systematically investigated.
Conclusion
In summary, our study demonstrates that T1 and T2 mapping techniques not only noninvasively assess renal injury associated with chronic liver disease but also enable dynamic monitoring of the therapeutic effects of BMSC interventions on renal injury.
Funding Statement
This work was supported by the National Natural Science Foundation of China (Grant No. 82271971) under the Ministry of Science and Technology of the People’s Republic of China, the Natural Science Foundation of Tianjin Municipality (Grant No. 24JCQNJC01110) under the Tianjin Municipal Science and Technology Bureau, and the Tianjin Key Medical Discipline Construction Project (Grant No. TJYXZDXK-3-012B).
Acknowledgments: Author Chen Zhang is employee of Siemens. We sincerely appreciate the support provided by Siemens Healthineers in the application of renal T1 mapping and T2 mapping sequences in our experiments.
Conflicts of Interest: The authors declare that they have no conflicts of interest.
Supplementary Materials: Supplemental Fig. 1
Supplemental Table 1 Consistency analysis of T2 values in normal rat kidneys at different time points (1-week interval).
Supplemental Table 2 Consistency analysis of T1 values in normal rat kidneys at different time points (1-week interval).
Supplemental Table 3 Diagnostic performance of renal T1 and T2 value.
References
- 1.Pose E, Piano S, Juanola A, Ginès P. Hepatorenal syndrome in cirrhosis. Gastroenterology 2024; 166:588–604.e1. [DOI] [PubMed] [Google Scholar]
- 2.Angeli P, Gatta A, Caregaro L, et al. Tubular site of renal sodium retention in ascitic liver cirrhosis evaluated by lithium clearance. Eur J Clin Invest 1990; 20:111–117. [DOI] [PubMed] [Google Scholar]
- 3.Lee HA, Seo YS. Current knowledge about biomarkers of acute kidney injury in liver cirrhosis. Clin Mol Hepatol 2022; 28:31–46. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Wang Y, Wang B, Qin J, et al. Use of multiparametric MRI to noninvasively assess iodinated contrast-induced acute kidney injury. Magn Reson Imaging 2024; 114:110248. [DOI] [PubMed] [Google Scholar]
- 5.Chen J, Chen Q, Zhang J, et al. Value of T2 mapping in the dynamic evaluation of renal ischemia-reperfusion injury. Acad Radiol 2022; 29:376–381. [DOI] [PubMed] [Google Scholar]
- 6.Huang Z, Xia X, Liang Y, et al. Assessment and integration of multiparametric MRI for liver fibrosis staging in rat non-alcoholic steatohepatitis: Evaluation of diagnostic efficiency and model interpretation. Eur J Radiol 2025; 182:111821. [DOI] [PubMed] [Google Scholar]
- 7.de Boer A, Harteveld AA, Stemkens B, et al. Multiparametric renal MRI: An intrasubject test-retest repeatability study. J Magn Reson Imaging 2021; 53:859–873. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Bura A, Stonciute-Balniene G, Velickiene L, Bumblyte IA, Vaiciuniene R, Jankauskas A. The possibilities of multiparametric magnetic resonance imaging to reflect functional and structural graft changes 1 year after kidney transplantation. Medicina (Kaunas) 2025; 61:1268. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Henry R, Goetsch T, Brandhuber L, et al. MRI quantitative T1 and T2 mapping of the renal cortex: Assessment of normal values and potential usefulness for renal masses at 3 T. Eur J Radiol 2024; 181:111741. [DOI] [PubMed] [Google Scholar]
- 10.Ko SF, Yip HK, Zhen YY, et al. Severe bilateral ischemic-reperfusion renal injury: Hyperacute and acute changes in apparent diffusion coefficient, T1, and T2 mapping with immunohistochemical correlations. Sci Rep 2017; 7:1725. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Ko SF, Yip HK, Zhen YY, et al. Renal damages in deoxycorticosterone acetate-salt hypertensive rats: Assessment with diffusion tensor imaging and T2-mapping. Mol Imaging Biol 2020; 22:94–104. [DOI] [PubMed] [Google Scholar]
- 12.Hueper K, Peperhove M, Rong S, et al. T1-mapping for assessment of ischemia-induced acute kidney injury and prediction of chronic kidney disease in mice. Eur Radiol 2014; 24:2252–2260. [DOI] [PubMed] [Google Scholar]
- 13.Imafuku A, Oka M, Miyabe Y, Sekiya S, Nitta K, Shimizu T. Rat mesenchymal stromal cell sheets suppress renal fibrosis via microvascular protection. Stem Cells Transl Med 2019; 8:1330–1341. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Zou X, Jiang K, Puranik AS, et al. Targeting murine mesenchymal stem cells to kidney injury molecule-1 improves their therapeutic efficacy in chronic ischemic kidney injury. Stem Cells Transl Med 2018; 7:394–403. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Wanyan P, Wang X, Li N, Huang Y, She Y, Zhang L. Mesenchymal stem cells therapy for acute kidney injury: A systematic review with meta-analysis based on rat model. Front Pharmacol 2023; 14:1099056. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Wang X, Xie S, Qiu C, et al. Use of intravoxel incoherent motion diffusion-weighted imaging to assess mesenchymal stromal cells promoting liver regeneration in a rat model. Acad Radiol 2024; 31:4955–4964. [DOI] [PubMed] [Google Scholar]
- 17.Xiao Z, Zhong J, Zhong L, et al. The prognostic value of myocardial salvage index by cardiac magnetic resonance in ST-segment elevation myocardial infarction patients: A systematic review and meta-analysis. Eur Radiol 2023; 33:8214–8225. [DOI] [PubMed] [Google Scholar]
- 18.Bulluck H, Bryant JA, Lim MX, et al. Full left ventricular coverage is essential for the accurate quantification of the area-at-risk by T1 and T2 mapping. Sci Rep 2017; 7:4871. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Zhu Y, Dong C, Xu Z, et al. Human umbilical cord mesenchymal stem cells alleviate diabetic nephropathy by inhibiting ferroptosis via the JNK/KEAP1/NRF2 signaling pathway. Antioxid Redox Signal 2025; 42:807–826. [DOI] [PubMed] [Google Scholar]
- 20.Wang Y, Lu D, Lv S, Liu X, Liu G. Mesenchymal stem cell-derived exosomes ameliorate diabetic kidney disease through NOD2 signaling pathway. Ren Fail 2024; 46:2381597. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Shah B, Anderson SW, Scalera J, Jara H, Soto JA. Quantitative MR imaging: Physical principles and sequence design in abdominal imaging. Radiographics 2011; 31:867–880. [DOI] [PubMed] [Google Scholar]
- 22.Kiricuta IC, Jr, Simplăceanu V. Tissue water content and nuclear magnetic resonance in normal and tumor tissues. Cancer Res 1975; 35:1164–1167. [PubMed] [Google Scholar]
- 23.de Miguel MH, Yeung HN, Goyal M, et al. Evaluation of quantitative magnetic resonance imaging as a noninvasive technique for measuring renal scarring in a rabbit model of antiglomerular basement membrane disease. J Am Soc Nephrol 1994; 4:1861–1868. [DOI] [PubMed] [Google Scholar]
- 24.Hueper K, Rong S, Gutberlet M, et al. T2 relaxation time and apparent diffusion coefficient for noninvasive assessment of renal pathology after acute kidney injury in mice: comparison with histopathology. Invest Radiol 2013; 48:834–842. [DOI] [PubMed] [Google Scholar]
- 25.Yang X, Luo Q, Wu Z, et al. Tanshinone IIA reduces tubulointerstitial fibrosis by suppressing GSDMD-mediated pyroptosis. Pharm Biol 2025; 63:364–373. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Yuan M, Yin Z, Wang Z, et al. Modification of MSCs with aHSCs-targeting peptide pPB for enhanced therapeutic efficacy in liver fibrosis. Biomaterials 2025; 321:123295. [DOI] [PubMed] [Google Scholar]
- 27.Feng XL, Qi WY, Xiao ZY, et al. Assessment of early anthracycline-induced cardiotoxicity and liver injury with T2 and T2* mapping in rabbit models. Eur Radiol 2024; 34:226–235. [DOI] [PubMed] [Google Scholar]