All animal experiments were approved by the Handong Global University Animal Care and Use Committee (Approval No. HGUIACUC20211214-18). C57BL/6 J (B6) female mice were purchased from Hyochang Science, Inc. (Daegu, South Korea) and maintained in a temperature and humidity-controlled environment on a 12 h dark/light cycle.

Female mice (6–9 weeks of age) were anesthetized with ketamine/xylazine (60 and 12 mg, respectively per kg body weight). Ischemia was induced by clamping the renal artery of the right kidney with nontraumatic microaneurysm clamps (Roboz Surgical Instrument, Gaithersburg, MD) for 45 min. During ischemia, body temperature was maintained at 36.8–37.2°C by placing mice on a heating pad. Mice were euthanized on days 0, 1, 6, or 20 following surgery and 40–50 mL of cold phosphate-buffered saline was administered through the left ventricle.

Kidney tissues were fixed in 4% paraformaldehyde and were embedded in paraffin. Serial 4-μm sections were stained with hematoxylin and eosin (H&E) and periodic acid-Schiff (PAS) to assess the renal injury, and von Kossa staining for calcium deposits. Fibrosis was assessed by Picosirius Red staining and quantifying collagen deposition (red staining) using ImageJ software (National Institute of Health, Bethesda, MD) in 5 randomly selected fields in the section of each group.

Total RNA was extracted using MiniBEST™ Universal RNA Extraction Kit (Takara Bio Inc., Shiga, Japan). Reverse transcription reaction was performed using Primescript™ 1st strand cDNA synthesis kit (Takara Bio Inc). Quantitative real-time PCR was performed using GoTaq™ qPCR Master Mix (Promega, Madison, WI) on a StepOnePlus™ Real-Time PCR system (Applied Biosystems, Foster City, CA). The amplification conditions were 95°C for 2 min, followed by 40 PCR cycles of 95°C for 15 s and 60°C for 1 min with SYBR green fluorescence detection. Primers are listed in Supplementary Table 1. The gene expression results were normalized to the expression of Gapdh, and the ΔΔCt method was used for calculating relative expression levels.

Western blot was performed as previously described (34). Kidney protein was extracted using a PRO-PREP™ protein extraction solution (iNtRON Biotechnology, Seongnam, South Korea), and 25 μg total protein was used for western blot. Proteins were separated by 10% SDS-PAGE and transferred to PVDF membranes. The membranes were blocked with 5% skim milk in TBS-T at room temperature for 1 h, then were probed with rabbit anti-mouse/human polyclonal F10/10a Ab or F13A (Invitrogen, Waltham, MA). Mouse monoclonal Direct-Blot HRP anti-GAPDH Ab (BioLegend, San Diego, CA) was used as a loading control. The proteins were visualized by ECL substrate solution and captured using a chemiluminescent imaging system (Azure 280, Azure Biosystems, Dublin, CA), and densitometric analyses were done using ImageJ software (National Institute of Mental Health, Bethesda, MD).

Fresh blood was collected in a BD Microtainer SST tube (BD Scientific, Franklin Lakes, NJ) and allowed to clot for a minimum of 30 min. Separated serum was frozen at -80’C until analysis. Serum F13a1 level was assayed using Mouse F13a1/F13A chain ELISA kit (ABclonal, Woburn, MA).

Kidney tissues were frozen in the OCT compound and stored at −80°C before processing. Serial 5-μm cryosections were stained for the presence of macrophages, using FITC-labeled rat anti-mouse CD68 or Alexa Fluor 488-labeled rat anti-mouse CD206 antibodies (BioLegend); and coagulation factors using FITC anti-mouse/human polyclonal F10/10a Ab or F13a (Invitrogen) followed by Alexa Fluor 568-labeled Goat anti-rabbit IgG (Invitrogen). Slides were mounted with a ProLong™ Gold Antifade Mountant with DAPI (Thermo Fisher Scientific, Waltham, MA). Images were taken using immunofluorescence microscopy (Carl Zeiss Axio Imager a2, Oberkochen, Germany) and processed with ImageJ software (National Institute of Mental Health, Bethesda, MD).

Kidney tissues were digested in RPMI medium, 0.1 mg/mL collagenase IV at 37°C for 1 h. Tissues were disaggregated by aspiration through 20G syringes and filtered through a 70-μm cell strainer. Cells were stained with PerCP/Cyanine5.5-labeled rat anti-mouse CD45 (clone: 30-F11), PE/Cyanine7-labeled rat anti-mouse/human CD11b (clone: M1/70), PE/Dazzle 594-labeled rat anti-mouse Ly6G (clone: 1A8), and FITC-labeled rat anti-mouse F4/80 (clone: BM8) antibodies by surface staining. Cells were washed with staining buffer (PBS, 0.5% w/v BSA, 0.01% w/v sodium azide) and fixed with 4% paraformaldehyde in PBS and permeabilized with 0.1% Triton X-100 (Sigma-Aldrich, St. Louis, MO) in PBS before staining with rabbit anti-mouse/human polyclonal F10/10a Ab or F13A (Invitrogen) followed by PE-labeled Donkey anti-rabbit IgG. Flow cytometry analysis was operated using an Attune NXT (Thermo Fisher Scientific) and the data were analyzed using FlowJo software v10.8 (BD Biosciences, Franklin Lakes, NJ). Unless otherwise stated, all antibodies were purchased from BioLegend.

Differential gene expression data were obtained from the Gene Expression Omnibus repository (GSE121410). Unbiased 2-dimensional hierarchical clustering and heatmap visualization of differential expressed coagulation factor genes were performed using an Array Studio 10 (OmicSoft, Cary, NC).

The results here are in whole or part based upon data generated by the Kidney Precision Medicine Project (KPMP): DK114886, DK114861, DK114866, DK114870, DK114908, DK114915, DK114926, DK114907, DK114920, DK114923, DK114933, and DK114937. Data were downloaded from https://www.kpmp.org on 8/29/2022 (35). Downstream analysis was performed using the R package Seurat (v4.0.6) and figures were generated with functionalities “Dimplot,” “FeaturePlot,” and “VlnPlot” (36) To reduce cluster numbers, clusters were renamed based on the KPMP study (37) using Python 3.

Mouse bone marrow cells were flushed from the femur and tibia and cultured in L929 cell-conditioned medium to separate adherent differentiated cells for 6 days. The media was changed every 2 days to remove nonadherent, and immature cells. To achieve polarization of BMMø, BMMø were stimulated with Lipopolysaccharides (LPS) (Sigma-Aldrich) or IL-4 and IL-13 (BioLegend) for 24 h. For calcium treatment, 50 mM calcium chloride (CaCl₂) was treated for 18 h and cell lysates were prepared.

Data represent the mean ± SEM prepared using GraphPad Prism 9.0 (GraphPad Software Inc., La Jolla, CA, United States). Statistical analyses were performed using the Mann–Whitney U test (one-tailed). The p values that were greater than 0.05 were considered significantly different. Statistically significant p values are denoted as *p < 0.05, **p < 0.01, and ***p < 0.001.

We hypothesized that kidney Mø express coagulation factors, which contribute to the provisional matrix formation, after an acute kidney injury (AKI). To test our hypothesis, we employed unilateral kidney ischemia–reperfusion (I/R) surgery, a murine model of sterile AKI (Figure 1A). Kidneys were then analyzed on days 1 (AKI), 6 (transition phase), and 20 (fibrosis). I/R kidneys were enlarged on day 1 and shrank until day 20 of I/R compared to contralateral (CL) kidneys (Figure 1B). AKI genes (Havcr and Lcn2) were elevated on days 1 and 6, and fibrosis genes (Col1a1, Col1a2, Col3a1, and Fn1) on days 6 and 20 of kidney I/R (Figures 1C,D, Supplementary Figure S1A). To verify that I/R induces kidney fibrosis on day 20, picrosirius red and PAS staining were performed. Collagen deposition (picrosirius red) and structural changes, such as tubular atrophy, and intratubular cast formation, were indicative of kidney fibrosis in I/R kidneys at day 20 (Figures 1E,F).

Next, we determined whether the expression of coagulation factors is increased within the kidney following kidney I/R and found that intrarenal F3, F7, and F10 transcripts were significantly increased until day 20 (Figure 1G). The expression of Intrarenal F13a1 transcript peaked in the transition phase (day 6) (Figure 1G). The protein levels of F10 and F13a1 showed an expression pattern corresponding to transcript data (Figures 1G,H, Supplementary Figure S1B). To determine whether the upregulated protein level of F13a1 within the kidney is derived from the tissue or circulation, we probed for the serum level of F13a1 at different time points after AKI (Figure 1I). Serum F13a1 level peaked on day 1 and immediately decreased to the basal level by day 6, indicating that upregulated intrarenal F13a1 on days 6 and 20 of I/R is a tissue-specific response.

Since F13a1 is known as a fibrin stabilizing factor, which crosslinks fibrin filaments to make fibrin polymer and stabilize clots, we investigated whether the fibrin matrix is present as the F13a1 level increases with H&E staining (Figure 1J). As the arrows point, the fibrin matrix was prominent on day 20 of the I/R group. Taken together, our data suggested that the coagulation factors including F13a1 are expressed by the kidney tissue following kidney I/R.

Our data indicated that the levels of coagulation factors (e.g., F10 and F13a1) are increased in the kidney after I/R. Next, we analyzed whether coagulation factors are expressed by kidney Mø found after I/R surgery.

We detected co-localization of Mø markers (CD68 or CD206) and F10 or F13a1 in the transition (day 6) and fibrosis phase (day 20) (Figure 2A, Supplementary Figure S2). In flow cytometry analysis, the mean fluorescence intensity (MFI) of F10 and F13a1 on Mø increased from day 1 to day 20 of I/R (Figure 2B).

Next, we questioned whether coagulation factors are expressed uniformly by all kidney Mø or only by specific kidney Mø subpopulations during kidney regeneration. Of note, I/R kidneys harbor a heterogeneous pool of Mø including infiltrating (Ly6C^high), MHC II⁺, and MHC II⁻ resident subpopulations (24). To answer our question, we probed for the expression of coagulation factors by different kidney Mø subpopulations using RNA sequencing (RNAseq) data, which were generated from kidneys at day 6 of I/R (GSE121410) (Figures 2C,D). To our surprise, unbiased hierarchical clustering analysis revealed that infiltrating and resident subpopulations express coagulation factors, which were distinct: Infiltrating Mø (M1-like) expressed coagulation factors driving the initiation (e.g., F7 and F10), whereas resident Mø (M2-like) produced factors responsible for the amplification of the coagulation cascade (e.g., F3 and F8) (Figures 2D,E, Supplementary Figure S3). In this analysis, we additionally found that the coagulation factor most strongly upregulated by resident Mø is F13a1, which catalyzes the last step of coagulation by crosslinking fibrin molecules to fibrin clots (Figures 2C,E). Next, we sought to verify our findings from mouse RNAseq data in humans. To this end, we examined the expression of coagulation factors by Mø using human kidney single-cell RNAseq data (Supplementary Figure S3). Our results indicated that F13A1 is upregulated in M2 Mø from AKI and CKD patients (Figures 3A,B) and the main source of F13A1 expression in human kidney patients is Mø (Figure 3C). In humans, we could not detect other coagulation factors expressed in Mø. It is conceivable that the transient expression of coagulation factors, combined with the inherent differences between human pathologies and animal models, could account for this discrepancy. Nevertheless, our data suggests that Mø-derived F13a1 may contribute to fibrinogenesis and in-tissue clotting and affect the development of kidney fibrosis.

Ca²⁺ plays an essential role in the coagulation cascade and is indispensable for the activation of several coagulation factors. In the conversion of prothrombin to thrombin, Ca²⁺ forms a complex with F10 and F5, forming the prothrombinase complex. We next probed for the presence of Ca²⁺ in I/R kidneys using von Kossa staining and found that Ca²⁺ is deposited adjacent to the proximal tubule and glomerulus in both transition (day 6) and fibrosis (day 20) phases, where Mø infiltrate (Figure 4A). This let us hypothesize that Ca²⁺ affects coagulation factor production in Mø.

To further explore the effects of Ca²⁺ on Mø, bone marrow-derived Mø (BMMø) were treated with CaCl₂ for 18 h (Figure 4B). In a Ca²⁺ concentration series, we found that BMMø produce coagulation factors in presence of Ca²⁺ in a dose-dependent manner. The expression of F5, F7, F10, and F13a1 in unpolarized (M0) BMMø increased dose-dependently until 50 mM, while F3 was not affected (Figure 4C). Also, M1 or M2 upregulated coagulation factors when treated with 50 mM Ca²⁺ (Figure 4D). F3 and F10 were significantly increased in M1 Mø, while F7 and F13a1 increased in M2 Mø upon Ca²⁺ treatment. Notably, the increase of F13a1 expression after Ca²⁺ treatment was most prominent compared to other coagulation factors. Taken together, we can conclude that Ca²⁺ induces the upregulation of coagulation factors, especially that of F13a1 in kidney Mø.