Amnion-derived mesenchymal stem cells attenuate anti-GBM glomerulonephritis via regulation of neutrophil CD44 expression

Abstract

Background

Mesenchymal stem cells (MSCs) possess immunomodulatory and tissue repair properties, and their therapeutic efficacy has been reported in models of glomerulonephritis. However, the clinical application of bone marrow-derived MSCs (BMSCs) is limited due to the invasive nature of bone marrow harvesting. In this study, we focused on amnion-derived MSCs (AMSCs), which can be obtained non-invasively after delivery, and investigated their therapeutic effects in a rat model of anti-glomerular basement membrane (GBM) nephritis.

Methods

In a rat model of anti-GBM nephritis, human AMSCs, BMSCs, or vehicle were intravenously administered. Evaluations included renal function, histology, intravital imaging of glomerular neutrophils, and flow cytometric analysis of leukocytes. In addition, we used an anti-CD44 neutralizing antibody in vivo to examine the therapeutic relevance of CD44 blockade, including its impact on neutrophil.

Results

The renoprotective effects observed in the AMSC-treated group were comparable to or greater than those in the BMSC-treated group. Furthermore, neutrophils exposed to AMSCs exhibited shorter dwell time in the glomeruli, and specifically in renal neutrophils, the expression of the adhesion molecule CD44 was reduced. In addition, administration of an anti-CD44 antibody produced similar renoprotective effects and reductions in neutrophil infiltration, suggesting that modulation of CD44 signaling contributes to the renoprotective effects of AMSCs.

Conclusions

Collectively, these findings indicate that AMSCs exert renoprotective effects comparable to or greater than those of BMSCs, and may represent a novel therapeutic approach for anti-GBM nephritis by specifically suppressing CD44 expression in neutrophils at sites of inflammation.

Graphical Abstract

Supplementary Information

The online version contains supplementary material available at https://doi.org/10.1186/s13287-026-05113-2.

Background

Mesenchymal stem cells (MSCs) are adult stem cells with the capacity to differentiate into various cell types, including osteoblasts, chondrocytes, and adipocytes. To define MSCs, the Mesenchymal and Tissue Stem Cell Committee of the International Society for Cellular Therapy (ISCT) has proposed the following minimal criteria [1]. First, MSCs must be plastic-adherent when maintained under standard culture conditions. Second, MSCs must express CD105, CD73, and CD90, and lack expression of CD45, CD34, CD14, or CD11b, CD79alpha or CD19, and HLA-DR surface molecules. Third, MSCs must differentiate into osteoblasts, adipocytes, and chondroblasts in vitro. By secreting a variety of growth factors, MSCs promote the proliferation and differentiation of resident cells at sites of injury. Consequently, the immunomodulatory and tissue repair properties of MSCs have attracted attention for their potential applications in regenerative medicine, autoimmune disorders, and inflammatory diseases [2].

MSCs can be isolated from several tissues, including bone marrow, skin, adipose tissue, and the amnion. Their characteristics—such as stem cell frequency, proliferative capacity, growth factor secretion, and differentiation potential—vary depending on the tissue of origin, and their properties can also be substantially influenced by the culture environment [3, 4]. Among them, bone marrow-derived MSCs (BMSCs) have been the most extensively studied and clinically applied, with demonstrated efficacy and safety in numerous preclinical and clinical studies, including those of glomerulonephritis [5, 6]. However, bone marrow harvesting is highly invasive, and obtaining sufficient amounts from healthy donors is challenging, highlighting the need for alternative, non-invasive sources of MSCs. Amnion-derived mesenchymal stem cells (AMSCs) can be obtained non-invasively from placental tissues that are otherwise discarded after delivery, with minimal ethical concerns, making them an attractive and abundant source of MSCs [7, 8]. Moreover, AMSCs have been reported to exhibit higher proliferative potential [9] and stronger immunomodulatory capacity than BMSCs [4], suggesting that they may possess advantageous properties for clinical application.

Anti-glomerular basement membrane (GBM) antibody disease (Goodpasture’s disease) is a rare autoimmune disorder in which autoantibodies targeting the α3 chain of type IV collagen deposit along the GBM and alveolar basement membranes, leading to rapidly progressive glomerulonephritis (RPGN) and diffuse alveolar hemorrhage [10, 11]. When both renal and pulmonary involvement is present, the condition is termed Goodpasture’s syndrome, whereas isolated renal involvement is referred to as anti-GBM glomerulonephritis. Histopathological features of anti-GBM glomerulonephritis include segmental or global fibrinoid necrosis, crescent formation, and disruption of Bowman’s capsule on light microscopy. Immunofluorescence reveals linear deposition of immunoglobulin G (IgG) and complement component 3 (C3) along glomerular capillary walls, while electron microscopy typically shows thickening and rupture of the GBM [12]. Treatment focuses on the rapid removal of pathogenic autoantibodies, usually through a combination of plasma exchange, corticosteroids, and immunosuppressants, aimed at suppressing autoantibody production and tissue inflammation. Early initiation of treatment often yields favorable renal outcomes; however, patients presenting with oliguria or anuria, extensive crescent formation, or dialysis-dependent renal failure generally have a poor prognosis [13]. In animal models of anti-GBM glomerulonephritis, intravenous injection of anti-GBM antibodies reproduces renal inflammation and pulmonary hemorrhage, with typical features including crescent formation, proteinuria, and elevations in serum creatinine and blood urea nitrogen (BUN) [14]. The pathogenesis of anti-GBM glomerulonephritis has been reported to involve diverse cellular components, including macrophages, lymphocytes, and intrinsic renal cells, as well as complex cytokine networks [15, 16]. In humoral immune-mediated injury, immunoglobulins activate complement and recruit macrophages and neutrophils via Fc receptors [17, 18]. In addition, cell-mediated immune responses involving CD4 + and CD8 + T cells can recruit macrophages through delayed-type hypersensitivity-like mechanisms, inducing glomerular injury even in the absence of immunoglobulins [19–21]. Despite these insights, the precise mechanisms underlying anti-GBM glomerulonephritis remain incompletely understood.

Several studies have reported that administration of BMSCs improves renal function, reduces proteinuria, and suppresses crescent formation in multiple models of glomerulonephritis, including anti-GBM glomerulonephritis [22, 23]. However, the effects of AMSCs on anti-GBM glomerulonephritis remain poorly characterized. The present study aimed to compare the effects of AMSC administration with those of BMSCs in a rat model of anti-GBM glomerulonephritis and to elucidate the therapeutic efficacy and underlying mechanisms of action in renal inflammation.

Methods

Animal experiments and procedures

Ethical statement and animals

All animal experimental protocols were approved by the Institutional Animal Ethics Committee of Nagoya University Graduate School of Medicine (approval number: M230165). All procedures were performed in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals and the ARRIVE guidelines 2.0. Female WKY/NCrj rats were purchased from Charles River, Inc. (Yokohama, Japan). At the time of the experiments, rats were 7–12 weeks old and weighed 140–190 g. Rats were housed in groups of 2–3 per cage under a 12-h light/dark cycle at 24 ± 1 °C with ad libitum access to a standard diet and water. No acclimatization period was applied. For each experimental set, rats were strictly age-matched across all treatment groups to ensure comparability.

Study design and sample size determination

The experimental unit was a single animal. A total of 200 rats were used across all experiments, conducted in at least two independent experimental repeats. The parameters for evaluating therapeutic efficacy were selected based on established protocols for this rat model [24]; changes in renal function (serum creatinine and BUN) and histological glomerular injury were defined as the main indicators for assessment. Sample sizes were determined a priori using G*Power software [25]. For the two-group comparisons, the required sample size was determined to be n = 6 per group based on a large anticipated effect size (Cohen’s d = 1.80), an α level of 0.05, and a power of 0.80. For the three-group comparisons, the required sample size was calculated to be n = 8 per group based on an effect size (Cohen’s f = 0.70), an α level of 0.05, and a power of 0.80. These parameters were selected based on preliminary experiments showing robust therapeutic efficacy, ensuring sufficient statistical power while minimizing the number of animals in accordance with the 3Rs principle (Replacement, Reduction, and Refinement). Actual group sizes ranged from n = 3 to n = 8, as specified in the figure legends. All data points represent independent biological replicates. No formal randomization or blinding was performed; rats were assigned to groups based on body weight. Experimenters were aware of group allocations. No inclusion or exclusion criteria were set, and no data were excluded from the analysis.

Induction of glomerulonephritis and MSC administration

Rapidly progressive glomerulonephritis was induced as previously described [24]. Briefly, a single dose of mouse monoclonal IgG clone TF78 (Shigei Medical Research Institute, Okayama, Japan), which specifically binds to α4 (IV) NC1 of the rat glomerular basement membrane, was diluted in GIT culture medium (FUJIFILM Wako Pure Chemical Corporation, Osaka, Japan) and intravenously administered via the tail vein at 100 µg/rat in a total volume of 500 µL. Healthy controls received an equal volume of GIT culture medium.

For MSC administration, unless otherwise specified, 2.0 × 10⁶ AMSCs or BMSCs were suspended in 2 mL of phosphate-buffered saline (PBS) containing 40 U/body of heparin and intravenously administered via the tail vein. Vehicle-treated animals consistently received the same volume of PBS (containing 40 U/body of heparin) via the same route to control for its anticoagulant effects. For the three-dose regimen, administrations were performed on days 0, 2, and 4. For the two-dose regimen, administrations were performed on days 0 and 4, and for the single-dose regimen, administration was performed on day 0. Cell preparation for injection was performed as described in the "Preparation of cells" section. For dose–response experiments, the administered cell numbers are specified in the corresponding figure legends.

Inhibition of CD44 by neutralizing antibody administration

For CD44 inhibition, anti-CD44 antibody (cat. no. GTX76383, GeneTex, Irvine, CA, USA) or isotype control antibody (cat. no. GTX18173, GeneTex) was administered via the tail vein once daily at a dose of 20 μg/rat in 1000 μL of PBS from day 0 to day 6.

Intravital imaging of glomerular neutrophil tracking

Tracking of intravenously administered neutrophils in rat kidneys was performed as previously described by Nishi et al. [26]. Rat neutrophils were isolated from bone marrow using histopaque gradient centrifugation, as described in detail below. The isolated neutrophils were labeled with 5 μM CFSE (Thermo Fisher Scientific, Waltham, MA, USA) or 20 μM CMF2HC (Thermo Fisher Scientific) in PBS for 30 min at 37 ℃. Cells were then washed twice and subsequently directly co-cultured with AMSCs in PBS containing 10% fetal bovine serum (FBS) for 2 h at 37 °C. The rat was anesthetized with a mixture of medetomidine, midazolam, and butorphanol, and injected with 0.01 mg/g body weight of rhodamine-70 k dextran to visualize the vasculature. The kidney was exteriorized with a lateral incision and extended over the imaging platform. Then, 5.0 × 10⁷ neutrophils in each color were injected into a rat via the tail vein without heparin. Real-time images were acquired using a Nikon A1RMP microscope (Nikon, Tokyo, Japan).

Physiological sampling, euthanasia, and perfusion fixation

Urine samples were collected on day 3 after nephritis induction to assess proteinuria. Blood samples were obtained on days 3 or day 7 to evaluate renal function, including serum creatinine and BUN. Serum creatinine and BUN were measured using standard automated biochemical analyzers by SRL, Inc. (Tokyo, Japan), and proteinuria was quantified according to standard laboratory protocols by Sanritsu Zelkova Inc. (Tokyo, Japan). The timing of sampling and tissue harvest is provided in the respective figure legends. At the designated endpoint (day 3 or day 7), rats were euthanized by exsanguination under deep anesthesia with isoflurane.

Perfusion fixation was performed specifically for the evaluation of fibrin clot formation and the localization of administered MSCs. For the fibrin clot evaluation, rats were administered anti-GBM antibodies and MSCs simultaneously, with or without heparin, and perfused one hour later. For the evaluation of MSC localization, rats received MSCs with 40U/body heparin on days 0 and 2, and were perfused on day 3. In both experiments, animals were perfused with 100 mL of PBS, followed by FM Fix (L) (FUJIFILM Wako Pure Chemical Corporation).

Analytical assays

Preparation of cells: MSCs, neutrophils, and CD4 + T cells

Human AMSCs were provided by Dr. Yamahara (Hyogo Medical University, Hyogo, Japan), and their preparation procedures have been described previously [27, 28]. Human BMSCs were purchased from Lonza (cat. no. PT-2501, Basel, Switzerland). For in vivo administration, cryopreserved MSCs (both AMSCs and BMSCs) were thawed, washed with PBS, and used immediately without further passage to ensure consistent cell quality.

Rat neutrophils used for in vivo cell tracking were isolated from femoral and tibial bone marrow, whereas those used for in vitro co-culture assays were independently isolated from peripheral blood and kidney tissues. For kidney-derived cells, tissues were digested with collagenase type I (Worthington Biochemical Corporation, Lakewood, NJ, USA) 1 mg/mL at 37 ℃ for 30 min prior to cell isolation. Thereafter, cells were processed as follows. Cells were suspended in PBS containing 2 mM EDTA and passed through a 100 μm cell strainer to obtain a single-cell suspension. Leukocytes were then separated by discontinuous density gradient centrifugation using Histopaque-1119 and Histopaque-1077 (Sigma-Aldrich), and contaminating red blood cells were lysed using Lysing Buffer (BD Biosciences) on ice for 5 min. The neutrophil-rich fraction was collected, and purity (> 90%) was confirmed by flow cytometry (see Supplemental Fig. 1 for the gating strategy).

CD4 + T cells were isolated from rat femoral and tibial bone marrow using the EasySep™ Rat CD4 + T Cell Isolation Kit (cat. no. 19642, STEMCELL Technologies, Vancouver, BC, Canada) according to the manufacturer’s instructions. Isolated cells were further purified by fluorescence-activated cell sorting (FACS) using a FACS SORP Aria II (BD Biosciences, San Jose, CA, USA) based on the markers listed in Supplemental Table 1.

In vitro co-culture assay

For in vitro co-culture assays of AMSCs and CD4 + T cells, AMSCs were co-cultured with purified CD4 + T cells at ratios of 1:20 and 1:1 in RPMI 1640 medium supplemented with 10% FBS and 1% penicillin–streptomycin (Thermo Fisher Scientific) for 39–48 h.

For in vitro co-culture assays of AMSCs and neutrophils, AMSCs were co-cultured with neutrophils at a ratio of 1:10 in RPMI 1640 medium supplemented with 10% FBS, 1% penicillin–streptomycin, and 1000 ng/mL lipopolysaccharide (LPS; cat. no. L4391, Sigma-Aldrich) for 16–20 h.

After co-culture, cells were collected and analyzed by flow cytometry.

Flow cytometry analysis of kidney and blood samples

Kidney and blood samples were processed for flow cytometry. Kidney tissues were enzymatically digested with 1 mg/mL collagenase type I (Worthington Biochemical Corporation) and 1 mg/mL DNase I (Merck, Darmstadt, Germany) at 37 °C for 30 min. The resulting cell suspensions were filtered through a 100 μm strainer, and red blood cells were lysed using BD Pharm Lyse (BD Biosciences). Cells were blocked with anti-CD32 (1:50; cat. no. 550271, BD Biosciences) and stained with antibodies listed in Supplemental Table 1.

Leukocyte subsets were analyzed on a FACS SORP Aria II (BD Biosciences) using FlowJo software (v10) (BD Biosciences), with gating shown in Supplemental Fig. 1. Absolute cell counts were determined using CountBright beads (cat. no. C36950, Thermo Fisher Scientific).

Evaluation of in vivo MSC localization

AMSCs or BMSCs were labeled with Vybrant™ DiD Cell-Labeling Solution (Thermo Fisher Scientific) at 37 °C for 30 min prior to administration. After washing with PBS, 1.0 × 10⁶ labeled cells were suspended in PBS containing 40 U/body of heparin and administered intravenously into anti-GBM nephritis rats on days 0 and 2. On day 3, sixty minutes prior to euthanasia, Alexa Fluor™ 568-conjugated Isolectin GS-IB4 (Thermo Fisher Scientific) was administered via the tail vein to label vascular endothelial cells in vivo. Following euthanasia, perfusion-fixed kidney tissues were harvested and sectioned. The sections were counterstained with DAPI (Vector Laboratories, Burlingame, CA, USA) and observed using a Nikon TiE-A1R confocal laser scanning microscope (Nikon). Excitation lasers at 405 nm, 561 nm, and 640 nm were used for image acquisition.

Histological, immunostaining, and quantitative analysis

Kidneys were fixed in formalin solution (Masked Formalin) at 4 °C for 48 h and paraffin-embedded. Sections were stained with Periodic Acid-Schiff (PAS) to evaluate glomerular and tubulointerstitial injury. Glomerular crescent formation was scored semiquantitatively by estimating the percentage of each glomerulus occupied by crescents: 0, no crescent; 1, 1–25%; 2, 26–50%; 3, 51–75%; and 4, 76–100%. At least 30 glomeruli per section were evaluated, and the mean score was calculated for each animal. Tubulointerstitial injury, defined by tubular dilation or atrophy, basement membrane denudation, or tubular necrosis, was scored on the same 0–4 scale across multiple cortical and medullary fields, and the mean injury score per section was calculated.

For immunohistochemistry, neutrophils were detected using naphthol-AS-D chloroacetate esterase staining (Sigma-Aldrich) as previously described [29].

Macrophages were detected using mouse anti-rat CD68 monoclonal IgG1 antibody (clone ED-1; BMA Biomedicals, Augst, Switzerland).

For immunofluorescence analysis of endothelial adhesion molecules, paraffin-embedded kidney sections were deparaffinized and rehydrated. Following antigen retrieval using citrate buffer and blocking non-specific signals using Blocking One (Nacalai Tesque, Kyoto, Japan), sections were incubated with goat anti-CD31 (1:20; cat. no. AF3628, R&D Systems, Minneapolis, MN, USA), rabbit anti-VCAM1 (1:200; cat. no. ab134047, Abcam, Cambridge, UK), and rabbit anti-rat ICAM1 (1:3000; cat. no. 80022-T60, Sino Biological, Beijing, China). After washing, sections were incubated with appropriate fluorophore-conjugated secondary antibodies and mounted with VECTASHIELD mounting medium (Vector Laboratories). To quantify VCAM1 and ICAM1 expression, images of at least 10 glomeruli from each rat’s kidney were acquired. Using Imaris software, CD31-positive blood vessels were segmented within the glomeruli, and the expression levels (mean fluorescence intensity) of VCAM1 or ICAM1 were assessed specifically in those vessels. The average values were calculated per animal.

For the evaluation of fibrin clot formation, lungs were fixed in formalin solution (Masked Formalin) at 4 °C for 48 h, cryoprotected in 30% sucrose, embedded in optimal cutting temperature (OCT) compound, and frozen. Frozen lung sections were stained with anti-fibrinogen antibodies (Agilent, Santa Clara, CA, USA) and mounted with VECTASHIELD mounting medium.

Statistical analyses

Statistical analyses were performed using the GraphPad Prism (GraphPad Software, San Diego, CA, USA). Data are presented as mean ± standard deviation (SD). Two-group comparisons were performed using Welch’s t-test. Comparisons among multiple groups were conducted using one-way ANOVA followed by Tukey’s post hoc test. Statistical significance was defined as P value < 0.05.

Results

Combining AMSC administration with heparin reduces mortality and pulmonary emboli in the anti-GBM glomerulonephritis model

Pulmonary embolism has been reported as a severe adverse effect of MSC administration [30]. To evaluate the in vivo effects of AMSC administration, particularly with regard to pulmonary embolism, we first examined the survival and lung pathology of rats. Anti-GBM glomerulonephritis was induced by administration of anti-GBM antibodies, and AMSCs were administered either on days 0, 2, and 4 (three doses) or on days 0 and 4 (two doses). In the group receiving 40 U/body of heparin simultaneously with AMSC administration, all rats survived until day 7, whereas approximately half of the rats in the no-heparin group died within 7 days (Fig. 1a).

Fig. 1.

Combination of AMSC administration with heparin reduces mortality and pulmonary emboli in the anti-GBM nephritis model. a Experimental scheme and survival curves for anti-GBM nephritis treated with a three intravenous AMSC administration (upper) or two administrations (lower), with or without heparin (40 U/rat; n = 4 per group). b Representative images of fibrinogen-stained lung sections (left) and quantification of the fibrinogen-positive area (right) according to heparin dose (0, 20, 40, and 80 U/rat; n = 4 per group). All data are presented as mean ± SD. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001 (one-way ANOVA)

Anti-GBM antibodies and AMSCs were then administered simultaneously with heparin at doses of 0, 20, 40, and 80 U/body, and fibrin levels in lung tissue were stained and quantified (Fig. 1b). Fibrin levels in the lungs decreased in a heparin dose-dependent manner. These results suggest that pulmonary embolism is a major cause of mortality following AMSC administration, and that co-administration of heparin reduces mortality by preventing pulmonary embolism.

AMSCs show dose-dependent therapeutic effects in the anti-GBM glomerulonephritis model

Next, we evaluated the therapeutic effects of AMSC administration in a rat model of anti-GBM glomerulonephritis. Because this model closely mimics human rapidly progressive glomerulonephritis during the early phase up to day 7, but transitions to a nephrotic syndrome-dominant phenotype thereafter, we limited our evaluation to day 7 to ensure clinical relevance [24]. Based on preliminary experiments (Supplemental Fig. 2), a dose of 2 × 10⁶ AMSCs per administration was selected for further evaluation. Rats receiving three doses of AMSCs after anti-GBM antibody administration exhibited decreased serum creatinine and BUN levels. Histological analysis also revealed a significant reduction in crescent scores in this group (Fig. 2a). Rats receiving two doses of AMSCs also showed reductions in serum creatinine and BUN, and improvement in crescent scores, although the effects were less pronounced compared with the three-dose group (Fig. 2b). In contrast, rats receiving a single AMSC dose showed no improvement in serum creatinine or BUN, and crescent scores were comparable to those in the vehicle group (Fig. 2c). These results indicate that AMSCs exert dose-dependent therapeutic effects in the anti-GBM glomerulonephritis model.

Fig. 2.

Human AMSCs exhibit dose-dependent therapeutic effects in the anti-GBM nephritis model. a–c Experimental schemes showing induction of anti-GBM nephritis followed by intravenous administration of AMSCs: three doses on days 0, 2, and 4 (a), two doses on days 0 and 4 (b), or a single dose on day 0 (c) (upper left). Serum creatinine and BUN on day 7 (upper right). Representative PAS-stained kidney sections and crescent scores on day 7 (bottom). Control indicates healthy rats without nephritis induction; vehicle indicates nephritis rats receiving vehicle only; AMSC indicates nephritis rats treated with AMSCs. All injections were performed with heparin (40 U/rat). All data are presented as mean ± SD for the control (n = 3), vehicle (n = 6), and AMSC (n = 6) groups. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001 (Welch’s t-test). Scale bars, 50 µm

AMSCs exhibit greater therapeutic effects than BMSCs in anti-GBM glomerulonephritis model rats

Next, we compared the therapeutic effects of BMSCs and AMSCs. Anti-GBM glomerulonephritis model rats received BMSCs or AMSCs on days 0, 2, and 4, and serum creatinine, BUN, and proteinuria were quantified on days 3 and 7 (Fig. 3a). Administration of MSCs resulted in significant reductions in serum creatinine and BUN, with a trend toward suppression of proteinuria; these improvements were particularly pronounced in the AMSC-treated group (Fig. 3b). Histological analysis revealed that crescent scores and tubular injury scores were also significantly reduced in the AMSC-treated group (Fig. 3c, d). Neutrophils and monocytes are among the earliest leukocyte subsets that adhere to deposited autoantibodies and are known to promote glomerular injury [31, 32]. Since AMSCs demonstrated therapeutic efficacy at a very early stage, these immune cells were histologically evaluated. Neutrophil infiltration in glomeruli and tubules was assessed using esterase staining. In glomeruli, the sites of autoantibody deposition, neutrophil numbers were decreased in the AMSC-treated group by day 3 (Fig. 3e), whereas no significant changes were observed in interstitial neutrophil counts (Fig. 3f). Infiltrating monocytes and macrophages were evaluated by ED1 staining. Similar to neutrophils, ED1-positive cells decreased in the glomeruli of the AMSC-treated group. Conversely, in the later phase of the disease, infiltration of ED1-positive cells into the interstitium and tubules was suppressed in the AMSC-treated group (Fig. 3g, h). In addition, DiD-labeled AMSCs and BMSCs were successfully detected in the kidney after administration (Supplemental Fig. 3), confirming that both cell types reached the target organ. These results demonstrate that MSC administration exerts both serological and histological improvements in anti-GBM glomerulonephritis, with therapeutic effects being more pronounced in the AMSC-treated group than in the BMSC-treated group.

Fig. 3.

AMSCs exhibit greater therapeutic effects in anti-GBM nephritis than BMSCs. a Experimental scheme showing induction of anti-GBM nephritis followed by three intravenous administrations of AMSC or BMSC on days 0, 2, and 4. b Serum creatinine, BUN, and proteinuria on days 3 and 7. c Representative PAS-stained glomerular sections on day 7 (left) and crescent scores on days 3 and 7 (right). Scale bars, 50 µm. d Representative PAS-stained tubulointerstitial sections on day 7 (left) and tubular injury scores on days 3 and 7 (right). Scale bars, 100 µm. e, f Representative images of glomerular (e) and tubulointerstitial (f) sections stained with esterase on day 3 (left) and quantification of esterase-positive cells in the glomerulus (e) and tubulointerstitial space (f) on days 3 and 7 (right). Scale bars, 20 µm (e), 100 μm (f). g, h Representative images of glomerular (g) and tubulointerstitial (h) sections stained with anti-ED1 antibody on day 3 (left) and quantification of ED1-positive cells in the glomerulus (g) and tubulointerstitial space (h) on day 3 and day 7 (right). Scale bars, 50 µm (g), 100 μm (h). Control indicates healthy rats without nephritis induction; vehicle indicates nephritis rats receiving vehicle only; AMSC indicates nephritis rats treated with AMSCs; BMSC indicates nephritis rats treated with BMSCs. All injections were performed with heparin (40 U/rat). All data are presented as mean ± SD. At day 3, n = 3–4 for the control group and n = 8 for the other groups; at day 7, n = 3 for the control group and n = 6–8 for the other groups. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001 (one-way ANOVA)

AMSC treatment modulates neutrophils, macrophages, and Tregs in the kidney and blood

To investigate the changes in renal leukocyte subsets after MSC treatment in a rat model of anti-GBM glomerulonephritis, we analyzed leukocyte subsets in the kidney by flow cytometry (Fig. 4a–c, Supplemental Figs. 1a, b, 4a). In this flow panel, T cells, CD4 + T cells, CD8 + T cells, Tregs, B cells, neutrophils, M1 macrophages, and M2 macrophages were identified. To provide a comprehensive picture of the changes in leukocyte subsets caused by the administration of MSCs, flow cytometry results were presented using t-distributed stochastic neighbor embedding (tSNE) (Fig. 4a, b). We assessed changes in the fraction of each leukocyte population to evaluate qualitative changes in the overall leukocyte population and further evaluated cell number quantitatively to assess the impact of these populations on the kidneys. Leukocyte subsets in peripheral blood were also examined in rats receiving BMSCs or AMSCs (Fig. 4d–f, Supplemental Fig. 4b).

Fig. 4.

AMSC treatment reduces renal neutrophils and increases Tregs in the kidney and blood. a, b, d, e T-distributed stochastic neighbor embedding (tSNE) visualization of flow cytometry data for leukocyte and T cell subsets in the kidney (a, b) and blood (d, e) on day 7. Data are presented as color-coded population overlays (left) and cell density maps (right) in tSNE space. c, f Frequencies of renal leukocyte subsets among CD45 + cells (neutrophils, M1 macrophages, M2 macrophages, and Tregs) and the M1/M2 ratio (upper), and absolute numbers of each subset (lower) in the kidney (c) and blood (f) on days 3 and 7 (day 3, n = 8 per group; day 7, n = 6–8 per group). All data are presented as mean ± SD. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001 (one-way ANOVA)

In the kidney of the AMSC-treated group, the significant increase induced by AMSC treatment was observed in the frequency of M2 macrophages at an early stage and regulatory T cells (Tregs) at a late phase. Similarly, in the peripheral blood, the proportion of Tregs was increased compared with controls. Furthermore, an in vitro co-culture system of AMSCs with CD4 + T cells demonstrated an increased proportion of Tregs (Supplemental Fig. 5), suggesting that AMSCs may directly promote Treg differentiation.

In contrast, there was a significant reduction in the proportion of neutrophils in the kidney of AMSC-treated group, whereas no substantial changes in neutrophil proportions were observed in the peripheral blood. These results suggest that the reduction in renal neutrophil proportions may contribute to the therapeutic effects of AMSCs in anti-GBM glomerulonephritis.

AMSCs reduce neutrophil adhesion to glomeruli by downregulating CD44 expression

Next, we investigated the mechanism underlying the reduction of neutrophils in the kidney. Histological analysis revealed that AMSCs reduced the number of neutrophils within the glomeruli but did not affect the number of neutrophils in the interstitium (Fig. 3e and f). Therefore, we focused on the glomeruli as the site for evaluating neutrophil dynamics. Using the method of two-photon intravital microscopy previously reported [26, 32], we determined neutrophil behavior in the glomerulus under inflammatory conditions. Neutrophils were isolated from the bone marrow of WKY/NCrj rats and labeled with CMF2HC or CFSE. To eliminate potential bias due to fluorescent properties, the dyes used for neutrophils co-cultured with AMSCs were alternated for each rat. After 2 h of co-culture, the labeled neutrophils and AMSCs were injected into anti-GBM glomerulonephritis model rats, followed by dextran injection to label the vasculature (Fig. 5a). Measurement of the dwell time of neutrophils in the glomeruli revealed that neutrophils co-cultured with AMSCs exhibited a significantly shorter dwell time compared with neutrophils not co-cultured with AMSCs, indicating that AMSCs directly inhibit neutrophil adhesion in the glomeruli (Fig. 5b). This evaluation system, which assesses differences in dwell time between neutrophils stained with two distinct colors within the same glomerulus, makes it possible to evaluate differences in the adhesion of individual neutrophils under the same inflammatory conditions. Therefore, we evaluated changes in the expression of adhesion molecules on neutrophils induced by AMSCs. Flow cytometric analysis was performed to assess the expression of adhesion molecules on neutrophils in the kidneys (Fig. 5c) and peripheral blood (Fig. 5d) of anti-GBM glomerulonephritis model rats treated with AMSCs or BMSCs. Administration of AMSCs reduced the expression of CD44 in neutrophils in the kidney, whereas no changes were observed in VLA4, CD11b, LFA1, or PSGL1 (Fig. 5c). In contrast, there was no significant difference in CD44 expression on neutrophils in the peripheral blood (Fig. 5d). In addition, to evaluate the effect on subsequent stable adhesion of neutrophils, immunofluorescence analysis of kidney sections demonstrated that ICAM1 expression in glomerular endothelial cells was decreased in AMSC-treated rats (Supplemental Fig. 6), suggesting that AMSCs also modulate endothelial adhesion properties. Consistent with these findings, in an in vitro co-culture system of AMSCs with rat neutrophils, CD44 expression was reduced in kidney-derived neutrophils but remained unchanged in peripheral blood-derived neutrophils (Supplemental Figs. 1c, 7). Taken together, these results indicate that AMSC treatment attenuates neutrophil recruitment to the glomeruli by downregulating CD44 expression on neutrophils and reducing endothelial ICAM1 expression, thereby suppressing neutrophil adhesion in a site-specific manner.

Fig. 5.

AMSCs reduce neutrophil adhesion and decrease CD44 expression. a Representative in vivo image of neutrophils adhering to glomeruli. A corresponding time-lapse movie is provided as Supplemental movie. b Distribution of neutrophil dwell time on glomeruli. Left panels show frequency distributions for AMSC-untreated (green) and AMSC-treated (red) groups. Right panel shows a cumulative percentile plot comparing the two groups. c, d Expression of adhesion molecules on neutrophils in the kidney (c) and blood (d) on day 3 of nephritis (n = 4–6 per group). Upper panels show histograms normalized to mode, and lower panels show quantification of mean fluorescence intensity (MFI) for CD44, VLA4, CD11b, LFA1, and PSGL1. All data are presented as mean ± SD. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001 (one-way ANOVA)

Administration of anti-CD44 antibodies ameliorates nephritis and reduces neutrophils in the kidney

We next evaluated the therapeutic effects of CD44 inhibition by administering anti-CD44 antibodies to anti-GBM glomerulonephritis rats (Fig. 6a). CD44 inhibition significantly reduced serum creatinine, BUN, proteinuria, and crescent scores. Moreover, histological analysis showed that the number of neutrophils in the glomeruli was significantly decreased in the anti-CD44 antibody-treated group (Fig. 6b), suggesting that CD44 inhibition suppresses neutrophil adhesion to the glomeruli. Furthermore, flow cytometric analysis after CD44 inhibition showed that only neutrophils were significantly altered in the kidney. In anti-CD44 antibody-treated rats, the proportion of neutrophils decreased in the kidney but increased in the blood (Fig. 6c, d, Supplemental Fig. 8), suggesting that suppression of neutrophil trapping in the glomerulus increased the number of neutrophils circulating in the blood.

Fig. 6.

Administration of anti-CD44 antibodies ameliorates nephritis and reduces neutrophils in the kidney. a Experimental scheme showing induction of anti-GBM nephritis followed by intravenous administration of anti-CD44 antibodies or isotype control antibodies from day 0 to day 6 (upper panel). Serum creatinine, BUN, proteinuria, and crescent score on day 7 are shown (lower panels). b Representative images of glomerular sections stained with esterase on day 3 (left) and quantification of esterase-positive cells in the glomerulus (right). c, d Frequencies of leukocyte subsets, including neutrophils, M1 and M2 macrophages, the M1/M2 ratio, and Tregs, in the kidney (c) and blood (d) on day 3 of nephritis. All data are presented as mean ± SD. n = 6 per group. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001 (Welch’s t-test)

Discussion

In this study, administration of AMSCs to rats with anti-GBM nephritis demonstrated dose-dependent renoprotective effects, including reductions in serum creatinine and BUN, improvements in crescent score and tubular injury score, and decreases in the numbers of neutrophils and macrophages within glomeruli and tubules. The renoprotective effects observed in the AMSC-treated group were comparable to, or even greater than, those of the BMSC-treated group. In particular, the AMSC group exhibited a significant reduction in renal neutrophil counts, which was attributed to decreased expression of CD44, an adhesion molecule on neutrophils, thereby shortening their dwell time in the glomeruli. Furthermore, the fact that inhibition of CD44 with an anti-CD44 antibody reproduced similar renoprotective effects suggests that regulation of neutrophil dynamics via CD44 plays an important role in the therapeutic effects of AMSC administration on anti-GBM nephritis.

Previous reports have shown that BMSCs improve renal function and suppress crescent formation in various models of nephritis [23, 33–38]. Accumulating evidence indicates that the renoprotective actions of MSCs are not mediated by differentiation into resident glomerular cells, but rather by paracrine effects and immunomodulatory signaling [24, 26, 39–44]. Indeed, MSCs have been shown to express growth factors such as hepatocyte growth factor (HGF), vascular endothelial growth factor (VEGF), and insulin-like growth factor-1 (IGF-1), and to contribute to renal protection by modulating inflammatory cytokine expression [45, 46]. In contrast, Tsuda et al. reported that fetal membrane-derived MSCs (FM-MSCs) improved renal injury when administered to rats with anti-Thy1 nephritis, but without significant induction of VEGF or HGF in the kidney after FM-MSC transplantation [47]. In the present study, we demonstrated that AMSCs may exert renoprotective effects equivalent to or greater than those of BMSCs, but the underlying mechanisms of nephritis improvement may differ. Specifically, whereas paracrine actions involving VEGF and other factors appear to be central in the case of BMSCs, AMSCs and FM-MSCs may primarily act by suppressing neutrophil adhesion and accumulation in the glomeruli through downregulation of CD44 expression, thereby reducing tissue injury.

CD44 is a cell surface glycoprotein that plays essential roles in cell proliferation, adhesion, migration, angiogenesis, inflammation, and cytoskeletal reorganization [48]. Furthermore, CD44 has been reported in both human and experimental models of IgA nephropathy, focal segmental glomerulosclerosis, and lupus nephritis to be associated with renal dysfunction, proteinuria, and glomerular and interstitial injury [49–52]. Regarding the relationship between anti-GBM nephritis and CD44, Jun et al. reported that in a rat model of anti-GBM nephritis, CD44 expression was markedly upregulated in infiltrating monocytes and lymphocytes, fibrotic regions, and damaged cortical tubules as the disease progressed [53]. In this study, we elucidated the mechanism by which AMSCs suppress neutrophil migration to the glomerulus and actively modulate the local immune response. In our in vitro co-culture systems, we observed that AMSCs not only directly downregulate CD44 expression specifically in kidney-derived neutrophils (Supplemental Fig. 7), but also significantly promote the induction of Tregs from purified CD4 + T cells (Supplemental Fig. 5). To our knowledge, this is the first report demonstrating that MSCs can directly regulate CD44 expression on neutrophils. This finding suggests that the reduced CD44 expression observed in vivo is not merely a result of the selective retention of CD44-high neutrophils, but rather an active immunomodulatory effect of AMSCs. This selective regulation suggests that AMSCs act in an inflammation site-specific manner, supporting their potential application as a novel therapeutic strategy.

Moreover, the local adhesion environment within the glomerulus was also altered by AMSC treatment. In the kidney tissues of nephritic rats treated with AMSCs, ICAM1 expression on glomerular endothelial cells was significantly suppressed, whereas VCAM1 expression remained unchanged (Supplemental Fig. 6). These coordinated effects—downregulating CD44 on neutrophils while simultaneously suppressing ICAM1 on endothelial cells— suggest a synergistic mechanism by which AMSCs prevent excessive neutrophil retention and subsequent tissue damage in the inflamed glomerulus.

However, we acknowledge several limitations in the present study. First, the precise molecular mechanisms underlying these immunomodulatory properties remain to be fully elucidated. Various soluble factors and extracellular vesicles secreted by MSCs have been reported to be broadly involved in neutrophil modulation and Treg induction in vitro [54–56]. Identifying the specific molecules responsible for the AMSC-mediated downregulation of CD44 and the subsequent increase in the Treg population represents an important challenge for future research. Second, neutrophil-specific CD44 inhibition was not performed, and thus it could not be directly demonstrated that the renoprotective effects of AMSC are mediated via neutrophil CD44. Future studies are required to establish a neutrophil-specific CD44 inhibition model and to characterize its functional effects. Third, in this study, human AMSCs were administered to immunocompetent rats, which introduces a xenogeneic context. Therefore, we cannot entirely exclude the possibility that some of the observed changes were influenced by immune responses against human antigens. However, in our previous studies using rat-derived MSCs in rat models of crescentic glomerulonephritis [38], we also observed comparable renoprotective and immunomodulatory effects. This suggests that the therapeutic effects observed in the present study are unlikely to be explained solely by xenogeneic immune responses. Even so, further studies are warranted to better assess their potential contribution.

Finally, pulmonary embolism has been reported as a risk associated with MSC administration [30]. In the present study, however, co-administration of heparin markedly reduced mortality. This represents an important finding for ensuring the safety of AMSC therapy in future clinical applications.

Conclusion

In conclusion, AMSCs exhibited renoprotective effects in a rat model of anti-GBM nephritis equivalent to or greater than those of BMSCs, with evidence suggesting that modulation of neutrophil dynamics through CD44 contributes to their mechanism of action. As an abundant and noninvasively obtainable cell source, AMSCs hold promise as a novel therapeutic option for nephritis, including anti-GBM nephritis.

Abbreviation

MSCs

Mesenchymal stem cells

AMSCs

Amnion-derived mesenchymal stem cells

BMSCs

Bone marrow-derived mesenchymal stem cells

GBM

Glomerular basement membrane

BUN

Blood urea nitrogen

RPGN

Rapidly progressive glomerulonephritis

IgG

Immunoglobulin G

C3

Complement component 3

OCT

Optimal cutting temperature compound

PBS

Phosphate-buffered saline

FBS

Fetal bovine serum

tSNE

T-distributed stochastic neighbor embedding

HGF

Hepatocyte growth factor

VEGF

Vascular endothelial growth factor

IGF-1

Insulin-like growth factor-1

FM-MSCs

Fetal membrane-derived mesenchymal stem cells

PAS

Periodic Acid–Schiff

Authors contributions

The manuscript was written by TN, KuF., KaF. The research idea and the study design were conceptualized by TN, KaF. Data acquisition and analysis were performed by TN, KaF. All authors performed data interpretation. SM provided supervision or mentorship. All the authors contributed with important intellectual content during the writing of the manuscript and its revisions.

Funding

This study was supported by JSPS KAKENHI (Kazuhiro F.), JST FOREST Program (JPMJFR200WKazuhiro F.) and AMED Program for Accelerating Medical Research(JP256f0137010 Kazuhiro F).

Data availability

The datasets generated and analyzed during the current study are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

All animal experimental protocols were conducted in accordance with the guidelines of the Institutional Animal Ethics Committee of Nagoya University Graduate School of Medicine, and were performed under the approved project titled “Amnion-derived cell therapy for kidney injury and autoimmune diseases in rats and mice” (Approval number: M230165, Date of approval: March 6, 2023). Written informed consent was obtained from all participants or their legally authorized representatives for the use of AMSCs. The study was approved by the Ethics Committee of Hyogo College of Medicine (Approval number: 0325, Date of approval: August 7, 2024) for the project titled “Preclinical study for development of therapies using human fetal-derived hematopoietic and mesenchymal stem cells.”

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.