Activated platelets promote endothelial dysfunction and monocyte recruitment in early renal microvascular injury

Abstract

Chronic kidney disease (CKD) is characterized not only by progressive fibrosis but also by systemic endothelial dysfunction and inflammation. Platelets, traditionally recognized for their role in hemostasis, also serve as key modulators of endothelial activation and immune cell recruitment. Platelet activation is commonly observed in patients with CKD and contributes to the proinflammatory environment. Although platelet-endothelial interactions are well-characterized in cardiovascular disease, their role in renal endothelial dysfunction and inflammation remains poorly understood. To investigate this, we used the unilateral ureteral obstruction (UUO) model in mice, to examine how platelet activation influences endothelial responses and monocyte/macrophage recruitment in the early phase of renal fibrosis development. Platelet depletion reduced the number of infiltrating macrophages in kidney tissue, decreased expression of endothelial activation and inflammation markers, and preserved the peritubular capillary (PTC) integrity. Further in vitro studies using human umbilical vein endothelial cells (HUVECs) showed that activated platelets induced endothelial dysfunction and inflammation, in line with the in vivo findings. To recapitulate the vascular microenvironment, we performed a shear flow-based transmigration assay. Monocyte adhesion and transendothelial migration significantly increased when endothelial cells were pretreated with activated platelets compared to unstimulated controls. Moreover, the presence of platelets on the inflamed endothelium further enhanced monocyte migration, suggesting a synergistic effect in promoting immune cell recruitment. Collectively, our findings highlight that activated platelets contribute to endothelial dysfunction, inflammation, and monocyte infiltration in early kidney injury, suggesting their potential as a therapeutic target to mitigate microvascular injury and preserve renal vascular integrity in kidney disease.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12950-026-00491-z.

Introduction

Chronic kidney disease (CKD) currently affects approximately 10% of the global population, and is a major cause of morbidity and mortality worldwide. CKD is characterized by a gradual decline in kidney function due to interstitial fibrosis and tubular atrophy, which is generally preceded by persistent microvascular dysfunction and chronic inflammation [1–4].

Endothelial dysfunction is recognized as a central and early event in CKD, playing a critical role in the development of microvascular inflammation and capillary rarefaction. This endothelial injury often precedes overt fibrosis and contributes to progressive renal damage. Studies have shown that CKD patients exhibit increased endothelial injury and impaired repair capacity, even in the early stages of the condition. This condition occurs as a consequence of oxidative stress and uremic toxins, which cause endothelial damage [5]. Supporting this, a recent study by Short et al. [6] reported that elevated circulating levels of endothelial activation markers, were independently associated with CKD incidence. These findings suggest that endothelial dysfunction is not only caused by CKD but also contributes to its progression.

Despite this growing recognition, the precise mechanisms by which endothelial injury promotes CKD remain incompletely defined [2, 4, 7]. One potential contributor that has received increasing attention is platelet activity. Over the past few decades, platelets have been found to have extensive and versatile functions [8]. Beyond their traditional roles in hemostasis, platelets act as key modulators of inflammation and endothelial phenotype. They release proinflammatory mediators and interact with leukocytes and endothelial cells, facilitating leukocyte recruitment and transendothelial migration [8, 9]. In vascular beds outside the kidney, such as in cardiovascular and systemic inflammatory disease, these platelet–endothelial interactions are well-characterized [10, 11]. However, whether similar interactions drive microvascular inflammation and endothelial dysfunction in the kidney remains underexplored.

While studies on platelet function in CKD patients report varying results -ranging from decreased to elevated activity- a growing body of evidence highlights their involvement in CKD progression [12–16]. Elevated platelet activity markers have been observed in CKD patients [12–14], suggesting enhanced platelet involvement. Emerging research implicates platelet activation in driving chronic inflammation, oxidative stress, immune reactions, and fibrosis that associated with CKD progression [17]. We have recently demonstrated that activated platelets promote kidney injury and fibrosis in models of acute kidney injury and diabetic kidney disease, and that platelet inhibition or depletion mitigates disease progression [18, 19].

Several antiplatelet agents are routinely used in CKD patients for cardiovascular protection, yet the biological mechanisms through which platelet activation may contribute to CKD progression remain poorly understood [20–22]. Bridging this clinical observation with experimental evidence, we applied a platelet-depletion strategy in a CKD model as a proof-of-concept approach to clarify the role of platelets in renal fibrosis and disease progression. Our study aimed to investigate how platelet–endothelial interactions shape immune responses in the kidney. We employed the unilateral ureteral obstruction (UUO) model—a well-established murine model of early kidney injury that induces sustained endothelial stress—to examine the role of platelets in endothelial dysfunction and immune cell recruitment. By combining in vivo platelet depletion with in vitro assays using HUVECs stimulated by activated platelets, we aimed to elucidate the mechanistic link between platelet–endothelial interactions and monocyte-mediated inflammation under physiologically relevant flow conditions. Our findings indicate that platelet activation promotes endothelial dysfunction and facilitates monocyte transmigration, supporting the notion that platelet–endothelial interactions may offer a promising therapeutic strategy to attenuate early microvascular injury and inflammation in CKD.

Methods

Unilateral ureter obstruction (UUO) Animal Model

Fourteen 8-week-old male C57BL6/J mice (Charles River Laboratories, France) were randomized between platelet depletion (n = 8) or isotype control (n = 6) and were housed in IVC cages with a 12/12 light/dark cycle and were allowed food and water ad libitum, with 3–4 animals housed per cage. All experiments were conducted in the Animal Research Institute Amsterdam at the Academic Medical Center. Following acclimatization, all study animals underwent ureteral obstruction of the right kidney under generalized anesthesia by isoflurane (3–4% induction, 1.5–2.5% maintenance, 100% oxygen). An abdominal incision was made to reach ureter which was subsequently ligated twice. In order to provide analgesia prior to surgery (0.5 h) and after surgery (6 h), 0.1 mg/kg subcutaneous injections of buprenorphine (Temgesic, Schering-Plough, the Netherlands) were administered. Directly after surgery, animals received intraperitoneal injections of anti-GP1b-IgG for platelet depletion, or non-depleting control IgG (1 mg/kg/day, Emfret Analytic, Germany). Three days after UUO, mice were sacrificed by exsanguination under generalized anesthesia. Prior to kidney harvest, animals were perfused via the left ventricle with chilled phosphate-buffered saline (PBS) to remove circulating blood components. The obstructed right kidney was divided into two parts. One half was snap frozen in -80 °C whereas the other was fixed in formalin.

Enzyme-linked immunosorbent assay (ELISA)

Platelet factor-4 (PF4) was measured in plasma and kidney harvested at day 3 post-UUO using mouse CXCL4/PF4 ELISA kit (R&D system, USA) according to the manufacturer’s instructions.

Immunohistochemistry

Mouse kidneys were fixed in 10% formalin for 24 h and embedded in paraffin. To prepare samples for immunohistochemistry staining, 4-µm thick paraffin sections were deparaffinized in xylene and rehydrated through an alcohol series. 3% hydrogen peroxide (Sigma Aldrich) was used to block any endogenous peroxide activity for 20 min at room temperature (RT), after which slides were boiled in a pH 6.0 sodium citrate buffer for antigen retrieval. Slides were incubated overnight at 4 °C with rabbit monoclonal IgG anti-CD42b antibody (1: 200, Abcam, #Clone SP219), rat monoclonal IgG anti-F4/80 antibody (1:50, Bio-rad, #Clone CI: A3-1) and rat monoclonal IgG2a anti-CD34 antibody (1:1600, Cedarlane, #CL8927AP). After three separate PBS washing the tissue sections were subsequently incubated for 30 min at RT with biotinylated goat anti-rabbit secondary antibody (Bright Vision Immunologic, #DPVR110HRP). DAB was applied to visualize the antibody staining after 15 min at RT. The slides were then dried, mounted, and counterstained with hematoxylin.

Quantitative image analysis was performed using Fiji software (v1.53, NIH, USA) or QuPath (version 0.4.3). F4/80⁺ macrophage and CD34⁺ endothelial staining were quantified as the percentage of positively stained area across 10 randomly selected non-overlapping fields per section at 20× magnification, focusing on the renal cortex and corticomedullary junction.

ICAM-1 expression was assessed by immunohistochemistry specifically in peritubular capillaries (PTC) within the tubulointerstitial compartment. Staining was evaluated using a semiquantitative scoring system based on staining intensity and the extent of positive capillaries: 0 = no staining, 1 = weak, 2 = moderate, and 3 = strong. Scoring was performed by the pathologist (J.J.R) in a blinded manner and applied uniformly across all experimental groups.

Peritubular capillary (PTC) density was assessed using CD34 immunostaining and quantified as the number of CD34⁺ peritubular capillary profiles per mm² in the tubulointerstitial compartment of the renal cortex and corticomedullary junction.

Platelet and monocyte isolation

Human peripheral blood was drawn from healthy volunteers (n = 4). Blood samples were collected using Sarstedt S-Monovette 10 ml 9NC tubes. Whole blood was centrifuged at 180 RCF for 15 min at 20 °C to produce platelet-rich plasma (PRP). After addition of acid citrate dextrose (tri-Sodium citraat.2H₂O 0.085 mol/L, glucose 0.11 mol/L, citric acid.H2O 0,071 mol/L), PRP was centrifuged for 20 min at 20 °C at 200 RCF. Isolated platelets were centrifuged in 200 RCF for 20 min at 20 °C after being resuspended in buffer A (NaCl 137 mmol/L, NaHCO3 11.9 mmol/L, glucose 5.6 mmol/L, MgCl2.6H2O 1 mmol/L, KCL 2.61 mmol/L, EDTA 2 mmol/L). Prostaglandin E1/PGE1 (1 μm, Chem Cruz Biotecnology) was added to prevent platelet activation during centrifugation. The Millipore Guava EasyCyte HT Flow Cytometer (Merck Millipore) was used to count platelet numbers, after which they were resuspended in Buffer B (similar with Buffer A but without EDTA). In order to activate the platelets, we used TRAP-6 (15 μm, Bachem, Germany) to stimulate them for 15 min at RT. Monocytes were isolated from human peripheral blood using the MACS monocyte isolation kit (CD14 microbeads human, Miltenyi Biotec, #130-050-201) according to the manufacturer’s instructions (Miltenyi Biotec GmbH). Monocyte counts were determined using a hemocytometer cell counting.

Flow cytometry

Circulating platelets were quantified in whole blood using flow cytometry (FACS Calibur, Becton Dickinson, USA). Whole blood samples were stained with an APC-conjugated anti-CD61 antibody (BioLegend, USA) to identify intact platelets, independent of their activation status. Platelet counts (CD61⁺ events) were used to confirm the efficacy of platelet depletion.

Cell culture

Human umbilical vein endothelial cells (HUVECs) were cultured in ECM medium (Endothelial Cell Medium, Sciencell, #1001) supplemented with fetal bovine serum (FBS, Sciencell, # 0025), endothelial cell growth supplement (ECGS, Sciencell, #1052) and antibiotic solution (P/S, Sciencell, #0503) according to the manufacturer’s protocol and kept at 37 °C under 5% CO₂. Endothelial cells were used up to passage 7 and seeded at 50,000 cells/ml in 12-well plates or at 25,000 cells/channel in fibronectin-coated Ibidi µ-slides VI0.4 (Ibidi, #80606). Medium in 12-well plates was refreshed every 3 days, whereas medium in µ-slides was refreshed twice daily. Then, cells were treated with fresh platelets every 3 days for 5 days. HUVECs were incubated with either resting or activated platelets (HUVEC: platelet ratio of 1:50). Control condition consisted of HUVECs without platelet incubation.

RNA extraction, reverse transcription and quantitative real-time PCR (qPCR)

Following the manufacturer’s instructions, TRIzol Reagent (Invitrogen and Life Technologies, Breda, The Netherlands) was used to extract total RNA from frozen kidney tissue or cell cultures. Oligo-dt primers were used to generate cDNA from 1 µg of RNA. The LightCycler480 (Roche Diagnostics, Almere, The Netherlands) was used for qPCR analysis of the mRNA expression of associated genes. TBP (for the UUO mouse model gene), GAPDH expression (for the human gene) were used to normalize gene expression. The SYBR green PCR master mix (Thermo-fisher, The Netherlands), specific forward and reverse primers, and ddH2O constitute the real-time PCR mixture. SYBR green dye intensity was analyzes with linear regression analysis using LinReg PCR software. The primer pairs used are shown in Supplementary Table S1.

Monocyte transmigration assay under flow

For tracking of monocytes, cells were labeled with Vybrant™ DiD Cell-labelling solution (1:6000, Invitrogen, #V22887) for 20 min at RT. Labeled monocytes were centrifuged for 3 min at 300xg at RT to wash away residual labelling solution and resuspended in fresh HEPES+ (100 mL HEPES + 100 mg D-Glucose + 2.5 mL Albuman + 100 µL 1 M CaCl2). The cells were incubated for 15 min at 37 °C before being used in the flow assay.

Transmigration was assessed using an Ibidi parallel-plate flow chamber to mimic physiological shear stress (0.5 mL/min, ~ 0.8 dyne/cm²). Two experimental approaches were employed. In the first, endothelial monolayers were pre-stimulated with activated platelets for five days under static conditions to induce endothelial activation, after which labeled monocytes (600,000 cells) were perfused over the endothelial monolayer under flow; platelets were not present during the monocyte flow phase. In the second approach, HUVECs were stimulated with TNF-α (1 ng/mL, 6 h) to induce endothelial inflammation, followed by perfusion of activated platelets over the inflamed endothelium under flow, and subsequently monocyte perfusion at a platelet-to-monocyte ratio of 100:1 [23]. These conditions were compared with TNF-α-stimulated HUVECs without platelet perfusion and with unstimulated HUVECs. Resting platelets were not included in either flow experiment.

Flow assays, including the observation of monocytes adhesion and transmigration, were imaged using an Axiovert 200 M widefield microscope (Zeiss, Germany) for a total duration of 20 min, in combination with live-cell fluorescent tracking. Signal was detected with an AxioCam ICc 3 (Zeiss, Germany) camera. A tile-scan of 4 × 6 frames was collected to quantify total adhesion and transmigration numbers. Images were taken using Zeiss Zen Blue software. The tile-scan was stitched using Zen Black software, using the DIC channel for stitching. Monocyte adhesion and migration were calculated using Imaris software version 9.7.2.

Statistical analysis

All data sets were tested for their distribution prior to analyses. Normally distributed data sets were analyzed by unpaired Student’s T-test or one-way ANOVA; data sets without normal distribution were analyzed by Mann-Whitney U test or Kruskal-Wallis test, with appropriate post hoc tests. Results are presented as mean ± SEM (standard error of the mean), unless indicated otherwise. The threshold for significance was set at P < 0.05. All analyses were performed using GraphPad Prism version 10.2.0 (GraphPad Software, San Diego, USA).

Results

Platelet depletion attenuates macrophage recruitment in UUO Kidney

In order to investigate the role of platelets in mediating vascular responses —particularly endothelial dysfunction and inflammation—circulating platelets were depleted by administration of an anti-GPIbα antibody following UUO surgery. We have previously demonstrated robust platelet accumulation in the renal microvasculature after UUO [24]. We confirmed effective platelet depletion in this study by a marked reduction in circulating CD61⁺ platelets in whole blood, as assessed by flow cytometry (P < 0.01; Fig. 1A), as well as by significantly reduced levels of platelet factor 4 (PF4) in both blood and renal tissue compared with isotype-treated UUO mice (P < 0.01; Fig. 1A).

Fig. 1.

Platelet depletion reduces inflammatory chemokine expression and renal monocyte infiltration at day 3 of UUO. (A) Circulating platelets were quantified in whole blood by flow cytometry using CD61⁺ staining to identify platelets. Platelet-depleted UUO animals showed a marked reduction in CD61⁺ platelet counts compared with isotype-treated UUO controls, confirming effective depletion of circulating platelets. PF4 levels in blood plasma and kidney tissue were measured by ELISA. (B) Representative images show F4/80 (macrophage marker) staining in renal tissue sections from the control and platelet depletion groups. The graph shows the percentage of F4/80-positive area in 10 non-overlapping fields, quantified using FIJI image analysis software. Scale bar, 50 μm. (C) Relative mRNA expression levels of Adgre1 (F4/80), Ccl5 (RANTES) and Ccl2 (MCP-1). Data are presented as mean ± SEM. *P < 0.05, **P < 0.01, determined by unpaired Student’s t-test or Mann-Whitney test

We next assessed the impact of platelet depletion on inflammatory cell recruitment. Immunohistochemical analysis of F4/80 revealed a significant reduction in renal macrophage infiltration in platelet-depleted mice compared with UUO isotype controls (Fig. 1B). This decrease was accompanied by lower expression of the pro-inflammatory chemokines Ccl2 (MCP-1) and Ccl5 (RANTES), key mediators of monocyte/macrophage recruitment, as measured by qPCR (Fig. 1C). Together, these findings indicate that platelet depletion effectively reduces both platelet presence and activation in the UUO-injured kidney, and attenuates early macrophage recruitment and local inflammatory signaling.

Platelet depletion attenuates endothelial dysfunction and maintains peritubular capillary density

Vascular integrity is pivotal in regulating inflammatory responses and maintaining tissue homeostasis. To investigate the impact of platelet activation on endothelial cells, we conducted qPCR analysis to assess mRNA expression levels of endothelial dysfunction markers.

Platelet depletion led to a significant reduction in the expression of adhesion molecules, including Vcam1 (P < 0.01) and Icam1 (P < 0.05), while Sele (E-selectin) remained unchanged compared to the UUO isotype control (Fig. 2A). The gene encoding endothelial nitric oxide synthase (Nos3), which is crucial for nitric oxide production, tended to increase in the platelet-depleted group, potentially indicating improved endothelial function, whereas Nos2, associated with inflammatory nitric oxide production, was significantly decreased in the platelet depleted group (P < 0.05) (Fig. 2B). Inflammatory cytokines Il6 and Il1β were also significantly downregulated (P < 0.05), indicating attenuated inflammatory signaling in the absence of platelets (Fig. 2C). The expression of Thbd (thrombomodulin), an endothelial anticoagulant marker, was elevated under platelet depletion, though not significantly, suggesting a trend toward enhanced antithrombotic activity. Meanwhile, Serpine1 (also known as PAI-1), a key inhibitor of fibrinolysis and a marker of endothelial stress and prothrombotic state, showed no significant difference between groups (Fig. 2D).

Fig. 2.

Platelet depletion attenuates endothelial dysfunction and maintains capillary density in obstructive nephropathy. The relative mRNA expression levels of (A) endothelial adhesion molecules (Vcam1, Icam1, Sele), (B) nitric oxide metabolism markers (Nos2, Nos3), (C) pro-inflammatory cytokines (Il1b, Il6), and (D) coagulation-related markers (Serpine1, Thbd) were analyzed using qPCR. (E) Representative images show CD34 staining in renal tissue sections from the control and platelet depletion group. Graph showing the number of CD34-positive capillaries in 10 non-overlapping fields was quantified using QuPath image analysis software. Data are presented as mean ± SEM. *P < 0.05, **P < 0.01, determined by Unpaired student t-test or Mann-Whitney test. (F) Representative images of ICAM-1 staining in peritubular capillaries (PTC) of UUO kidneys. ICAM-1 expression was semiquantitatively scored (0–3) only in PTC endothelium. Positive staining was also observed in arterioles and tubular epithelial cells, but these compartments were not quantified. Data are presented as median with interquartile range (IQR) (p = 0.0506, Mann–Whitney test). *ns indicates not significant

Capillary rarefaction is one of the early events preceding kidney fibrosis. To assess peritubular capillary density, UUO kidneys were stained with CD34 to label endothelial cells. Platelet depletion resulted in significantly higher capillary density compared to untreated UUO mice (P < 0.01), demonstrating a protective effect of platelet depletion on microvascular integrity under sustained endothelial stress (Fig. 2E).

ICAM-1, a key endothelial adhesion molecule involved in leukocyte recruitment, was semiquantitatively assessed in peritubular capillaries (PTC). UUO isotype control kidneys showed a trend toward increased ICAM-1 expression in PTC compared with platelet-depleted animals, consistent with Icam1 mRNA, although the difference did not reach statistical significance (median 2 vs. 1; p = 0.0506, Mann–Whitney test, Fig. 2F). Notably, this trend toward reduced ICAM-1 expression in platelet-depleted kidneys coincided with increased PTC density, suggesting a relationship between preserved capillary integrity and lower endothelial activation. Representative images illustrate prominent ICAM-1 staining in UUO PTC and its attenuation following platelet depletion.

Platelet activation promotes endothelial dysfunction in HUVEC endothelial cells

To further investigate the effect of platelets on endothelial cells, we incubated HUVECs with either resting or activated human platelets for 5 days. We analyzed the relative mRNA expression levels of endothelial dysfunction markers. Significant increases were observed in VCAM1, ICAM1, and SELE, along with a reduction in NOS3, in the HUVEC cells treated with activated platelets compared to the control group (Fig. 3A-B). Consistent with our in vivo study, HUVECs treated with activated platelets induced a marked upregulation of proinflammatory cytokines and chemokines, including IL1β, IL6, TNF, CCL2, CCL5 and CXCL8 (IL8) (Fig. 3C-D). We also observed a significant decrease in THBD and an increase in SERPINE1 expression in response to platelet stimulation (Fig. 3E).

Fig. 3.

Platelet activation promotes endothelial dysfunction and inflammation in human endothelial cells. Relative mRNA expression levels were measured by quantitative PCR (qPCR) assay and normalized to the housekeeping gene GAPDH. HUVECs were either unstimulated (Control), co-cultured with resting platelets (Rest Plt), or co-cultured with activated platelets (Act Plt) for 5 days prior to RNA extraction. Panel A shows endothelial adhesion molecules (VCAM1, ICAM1, SELE), panel B shows the nitric oxide metabolism marker (NOS3), panel C depicts proinflammatory cytokines (IL1β, IL6, TNF), panel D illustrates proinflammatory chemokines involved in monocyte recruitment (CCL2, CCL5, CXCL8/IL8) and panel E presents coagulation-related markers (THBD, SERPINE1). Data are presented as mean ± SEM. *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001 determined by one-way ANOVA test or Kruskal-Wallis test. Abbreviations: Rest Plt = HUVECs co-cultured with resting platelets; Act Plt = HUVECs co-cultured with activated platelets

When exposed to resting platelets, we also observed an increase in markers associated with endothelial dysfunction and inflammation, although their levels were not as high as those in the group exposed to activated platelets and, in most cases, did not reach statistical significance.

Activated platelets facilitate monocyte transendothelial migration

Since platelet depletion reduced the recruitment of intra-renal macrophages (Fig. 1), we next explored the role of platelets in monocyte transmigration under physiologically relevant flow conditions. HUVEC monolayers were either pre-stimulated with activated platelets for 5 days (platelet pre-stimulation) or stimulated with TNF-α (1 ng/mL, 6 h) followed by perfusion of activated platelets (Fig. 4A). Freshly isolated human monocytes were subsequently perfused over the endothelial monolayers in an Ibidi flow chamber to mimic shear stress (~ 0.8 dyne/cm²).

Fig. 4.

Activated platelets enhance monocyte trans-endothelial migration. (A) Confluent HUVEC monolayers were stimulated with activated platelets, followed by perfusion of freshly isolated monocytes under physiological flow, to assess platelet-induced endothelial activation. In a second condition, endothelial monolayers were pretreated with TNF-α before platelet stimulation and monocyte perfusion, to examine the contribution of platelets to monocyte recruitment under dysfunctional endothelium. (B) HUVEC monolayers were treated with activated platelets for 5 days. Quantification of adherent and transmigrated monocytes shows that platelet stimulation significantly increased both monocyte adhesion and transmigration compared with unstimulated controls. (C) Monocyte adhesion and transmigration were assessed in TNF-α-stimulated HUVECs with or without prior perfusion of activated platelets. TNF-α increased both adhesion and transmigration, while addition of activated platelets significantly further enhanced transmigration but not adhesion. Minimal adhesion and transmigration occurred in the absence of TNF-α. (D) Representative images of adhered (white) and transmigrated (red) monocytes under each condition. Inserts indicate enlarged views of the areas highlighted by boxes. Data are presented as mean ± SEM. *P < 0.05, **P < 0.01, ***P < 0.001, determined by one-way ANOVA test or Kruskal-Wallis test. Abbreviations: Plt = platelets

In the platelet pre-stimulation condition, platelet-activated endothelial cells exhibited increased markers of dysfunction and inflammation compared with unstimulated HUVECs (Supplementary Fig S1). Functionally, this resulted in increased monocyte adhesion and transmigration, whereas minimal adhesion and migration were observed in HUVECs that were not pretreated with platelets (Fig. 4B). These findings indicate that platelet-induced endothelial activation creates a permissive environment for monocyte recruitment.

Under TNF-α-induced endothelial inflammation, perfusion of activated platelets further enhanced monocyte transmigration, while adhesion was only modestly increased and did not reach statistical significance (Fig. 4C–D). Minimal adhesion and transmigration occurred in the absence of TNF-α.

These observations indicate that endothelial activation is required for robust monocyte adhesion, while platelet exposure specifically amplifies transmigration. Notably, the effect on adhesion is partially attenuated under flow, likely due to the reduced stability and duration of platelet–endothelial and monocyte–endothelial interactions. Nevertheless, the cooperative effect of TNF-α and platelet signaling demonstrates additive enhancement of transmigration, reflecting physiologically relevant interactions under shear stress.

Discussion

Although CKD patients commonly receive antiplatelet therapy to reduce cardiovascular risk, the specific role of platelet activation in driving renal inflammation and endothelial dysfunction remains unclear [20–22]. Using a platelet-depleted CKD model and in vitro experiment, our study provides mechanistic evidence that platelets actively contribute to endothelial activation, immune cell recruitment, and rarefaction of peritubular capillary. Here, we discuss how these platelet-mediated processes promote renal inflammation and microvascular alterations, and their potential implications for therapeutic intervention.

The endothelium plays a central role in ensuring tissue perfusion by regulating vessel tone and diameter. It also acts as a selective barrier that modulates the exchange of molecules and immune cells between the blood and surrounding tissues. Under inflammatory or injurious conditions, endothelial cells shift toward an activated phenotype, characterized by increased expression of adhesion molecules and proinflammatory cytokines, promoting leukocyte recruitment. In contrast, under resting conditions, endothelial cells express inhibitors of platelet adhesion and coagulation, thus maintaining a non-thrombogenic surface [4, 25]. Upon vascular injury or inflammation, platelet activation facilitates adhesive interactions with leukocytes and the endothelium, notably through P-selectin–PSGL-1 engagement and stabilization via vWF-, integrin αIIbβ3-, ICAM-1– and fibrinogen-dependent mechanisms [9, 26, 27].

Here, we demonstrated that platelet activation contributes to endothelial dysfunction, inflammation and alteration in peritubular capillary density, all contributing to the pathogenesis of kidney fibrosis. Using a UUO model, platelet depletion reduced endothelial activation, inflammatory marker expression, and macrophage infiltration, while preserved PTC density. Furthermore, our in vitro assay showed that activated platelets were capable of inducing endothelial dysfunction characterized by increased expression of adhesion molecule markers, proinflammatory cytokines and chemokines, nitric oxide metabolism and altered coagulation-related markers in the endothelium. Flow-based transendothelial migration assays further provided functional evidence that monocyte adhesion and transmigration significantly increased in dysfunctional endothelium due to previous platelet stimulation, and that the presence of platelets on inflamed endothelium cells even further amplified monocyte transmigrating. These findings suggest that platelet-endothelial interactions may prime the endothelium to support immune cell recruitment, which may contribute to ongoing inflammation and microvascular damage in CKD.

We identified that platelet depletion in UUO was associated with lower PF4 levels in plasma and renal tissue, as well as lower renal expression of Ccl5 (RANTES) and Ccl2 (MCP-1). Platelet Factor 4 (PF4), known as the most abundant chemokine released upon platelet activation, plays a role in macrophages recruitment. CCL5 and CCL2 are pivotal in regulating the migration and localization of macrophages within the tissue [26, 28]. The presence of PF4 was found to enhance CCL5 (RANTES)-induced leukocyte arrest on activated endothelial cells [29].

While PF4 is derived from platelets, CCL2 and CCL5 are primarily produced by renal parenchymal cells and infiltrating immune cells rather than by platelets themselves. Publicly available single-cell RNA sequencing data from the Humphreys laboratory in a severe UUO model (e.g., day 14) indicate that Ccl2 expression is predominantly localized to failed-repair proximal tubular epithelial cells and macrophages, whereas Ccl5 displays a more heterogeneous cellular origin, including renal tubular, stromal, activated immune, and endothelial cells [30]. Despite the presence of endothelial cells in the dataset, indicated by detectable endothelial Ccl5 expression, their contribution to Ccl2 expression appears limited.

Therefore, the reduction in CCL2 and CCL5 observed upon platelet depletion is unlikely to reflect direct chemokine production by platelets. Instead, these findings suggest that platelets indirectly modulate renal chemokine expression by influencing the activation state of endothelial and other renal parenchymal cells. In support of this interpretation, our in vitro experiments demonstrate that exposure of endothelial cells to activated platelets induces endothelial dysfunction and a robust proinflammatory response, including the upregulation of CCL2 and CCL5. Together, these findings support the concept that platelets shape the inflammatory milieu of the injured kidney by regulating chemokine expression in resident renal and vascular cells rather than serving as a primary cellular source. This platelet-endothelial crosstalk likely contributes chemokine expression and creates a pro-recruitment microenvironment. We observed increased CXCL8/IL8 expression in HUVECs stimulated by activated platelets. IL-8, a chemokine primarily involved in neutrophil migration and activation, has also been reported to participate in monocyte recruitment and modulates macrophage polarization toward pro- or anti-inflammatory phenotypes, highlighting the broader impact of platelet-endothelial interactions on immune cell function [31, 32].

Platelet activation also modulated endothelial adhesion molecules. In vivo, platelet depletion reduced the expression of key adhesion molecules, such as Icam1 and Vcam1, whereas in vitro, activated platelets upregulated ICAM1, VCAM1 and E-selectin, promoting leukocyte rolling and firm adhesion, consistent with their roles in monocyte recruitment [33]. Resting platelets also led to increased markers of endothelial dysfunction and inflammation, although to a lesser extent compared to activated platelets. Resting platelets are usually quiescent but contain diverse molecules and granules in their cytoplasm that potentially be able to trigger the endothelium. Furthermore, during the 5-day co-culture, resting platelets may undergo partial activation due to mechanical stress, contact with endothelial-derived agonists, temperature fluctuations, or interaction with plastic surfaces, potentially contributing to subtle alterations in endothelial cells. These considerations highlight the dynamic nature of platelet-endothelial interactions and the need for careful interpretation of in vitro results.

We observed that proinflammatory cytokine markers including IL-1β, IL-6, and TNF-α were significantly upregulated in endothelial cells upon stimulation with activated platelets, while platelet depletion in UUO reduced these cytokines, supporting the role of platelets as upstream mediators of endothelial activation and inflammation [34]. To investigate the functional consequences of platelet-induced endothelial activation under physiologically relevant shear stress, we assessed monocyte recruitment in flow-based assays using two complementary experimental setups. In the first setup, platelet-primed endothelium significantly increased monocyte adhesion and transmigration compared with unstimulated endothelium, demonstrating that endothelial activation alone creates a permissive environment for monocyte recruitment. While shear stress partially limited the duration and stability of endothelial–monocyte interactions, transmigration remained robust. In the second setup, when activated platelets were applied to TNF-α–stimulated endothelium, monocyte recruitment was further enhanced, indicating an additive effect of platelet-derived signals on endothelial activation under pre-existing inflammatory conditions. Together, these results highlight the cooperative contribution of platelet priming and endothelial activation in driving monocyte recruitment under physiologically relevant flow, underscoring the in vivo relevance of platelet–endothelial crosstalk in renal inflammation.

In addition to promoting inflammation, activated platelets also influence endothelial pro- and anticoagulant balance. In our study, Platelet-stimulated HUVECs upregulated SERPINE1/PAI-1, a procoagulant factor, and downregulated thrombomodulin (THBD), an anticoagulant protein [25, 35]. In vivo, platelet depletion tended to restore Thbd expression, although without statistical significance, while Serpine1 remained unchanged. Taken together, these findings reinforce the concept that platelet activation contributes to endothelial dysfunction not only through inflammatory and adhesive pathways but also by disrupting the hemostatic balance toward a procoagulant phenotype.

Reduced peritubular capillary density is often referred to as capillary rarefaction in CKD models, which contributes to tissue hypoxia and fibrosis. Peritubular capillary (PTC) rarefaction is one of the earliest vascular changes observed in the progression of CKD. This condition is primarily driven by apoptosis and senescence of peritubular endothelial cells and impaired angiogenesis, predominantly mediated by TGF-β, which ultimately leads to capillary loss, tissue hypoxia, and fibrosis [36–39]. Our data suggest that platelet depletion alleviates endothelial dysfunction and restores of PTC density in UUO kidney, aligning with studies showing that both platelet depletion and P2Y12 receptor blockade reduce microvascular injury and endothelial loss [40]. Notably, this increase in capillary density was accompanied by a trend toward reduced ICAM-1 expression in peritubular capillaries, suggesting that improved endothelial preservation is associated with decreased endothelial activation and pro-adhesive signaling. Capillary rarefaction is also associated with decreased expression of endothelial nitric oxide synthase (NOS3 or eNOS) [36]. Our in vitro study demonstrated that stimulation of endothelial cells with activated platelets led to a reduction in NOS3 gene expression and in vivo, platelet depletion was associated with improved peritubular capillary density and a trend toward increased NOS3 expression, although the latter did not reach statistical significance. NOS3, derived nitric oxide plays a protective role against capillary rarefaction in CKD. In a study using the remnant kidney model, eNOS-deficient mice exhibited accelerated loss of glomerular and peritubular capillaries, driven by reduced endothelial cell proliferation and increased apoptosis, leading to enhanced inflammation, fibrosis, and progression of kidney injury [41]. Together, these data support a model in which platelet–endothelial interactions can influence microvascular health through modulation of NOS3 signaling.

In summary, our study demonstrates that activated platelets modulate endothelial dysfunction and inflammatory signaling, thereby facilitating monocyte transendothelial migration and contributing to renal microvascular alterations. Prior exposure of endothelial cells to activated platelets further enhances monocyte recruitment, whereas reduction of platelet availability within renal tissues attenuates endothelial activation and reduces macrophage infiltration. Collectively, these findings underscore the importance of platelet–endothelial interactions in renal inflammation and suggest that modulation of platelet activation may represent a potential therapeutic strategy to preserve endothelial integrity and limit early microvascular injury and consequent fibrosis development in chronic kidney disease.

Limitations of the study

We acknowledge several limitations of the present study. Isotype-treated UUO kidneys were used as the baseline comparator rather than non-obstructed (naive) kidneys, which may have provided additional physiological context. The cellular sources of CCL2 and CCL5 could not be definitively assigned, as gene expression analyses were performed on whole-kidney lysates, thereby precluding precise cell-specific attribution. Although precautions were taken to maintain platelets in a resting state during in vitro experiments, partial activation during handling or prolonged incubation cannot be entirely excluded and may have influenced endothelial responses. Circulating platelet depletion was confirmed by CD61 staining; however, comprehensive profiling of other circulating leukocyte populations was not performed. Finally, while monocyte adhesion and transmigration assays were conducted under flow conditions, endothelial activation was induced under static conditions prior to flow experiments, which may not fully recapitulate the dynamic in vivo microenvironment. Taken together, these considerations define the scope of the present study in examining platelet-mediated endothelial and inflammatory responses during early kidney injury.

Supplementary Information

Below is the link to the electronic supplementary material.

Abbreviations

CKD

Chronic kidney disease

UUO

Unilateral ureteral obstruction

PTC

Peritubular capillary

HUVECs

Human umbilical vein endothelial cells

IVC

Individually ventilated cages

RT

Room temperature

PBS

Phosphate-buffered saline

DAB

3,3′-Diaminobenzidine

FBS

Fetal bovine serum

ECGS

Endothelial cell growth supplement

PRP

Platelet-rich plasma

RT-qPCR

Reverse transcription–quantitative polymerase chain reaction

ELISA

Enzyme-linked immunosorbent assay

GP

Glycoprotein

vWF

Von Willebrand factor

Author contributions

UJR, SF, JDvB, AT and JJR designed the experimental strategy, interpreted the results. UJR, AT and JJR led the project and wrote the manuscript. MU performed animal experiments. UJR, MU, XZ, AMDvS, NC and LB performed laboratory experiments. UJR drafted the original manuscript. All authors read and edited the manuscript.

Funding

This research is supported by The Directorate General of Higher Education, Ministry of Education, Culture, Research, and Technology of Republic of Indonesia to UJR. JJR’s research was supported by the Netherlands Organization for Health Research and Development (ZonMw, grant no. 40-00703-97-12480). AT is supported by a NWO-FAPESP grant on healthy aging executed by Zonmw and a Junior Talent grant from Nierstichting.

Data availability

The authors confirm that the data supporting the findings of this study are available within the article and its Supplementary Material.

Declarations

Ethics approval and consent to participate

Animal studies were conducted with approval from the University of Amsterdam Animal Care and Use Committee (reference number DPA4) and in accordance with Dutch government legislation. All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards. The study was approved by the Medical Research Ethics Committees (MREC) of Amsterdam UMC (reference number 2024.0885).

Consent for publication

No applicable.

Competing interests

The authors declare no competing interests.