Elimination of the fibrinogen integrin αMβ2-binding motif improves renal pathology in mice with sickle cell anemia

Abstract

Sickle cell anemia (SCA) is caused by a point mutation in the β-globin gene that leads to devastating downstream consequences including chronic hemolytic anemia, episodic vascular occlusion, and cumulative organ damage resulting in death. SCA patients show coagulation activation and inflammation even in the absence of vascular occlusion. The coagulation factor fibrinogen is not only central to hemostasis but also plays important roles in pathologic inflammatory processes, in part by engaging neutrophils/macrophages through the α_Mβ₂ integrin receptor. To determine whether fibrin(ogen)-mediated inflammation is a driver of SCA-associated pathologies, hematopoietic stem cells from Berkeley sickle mice were transplanted into homozygous Fibγ^390-396A mice that express normal levels of a mutant form of fibrin(ogen) that does not engage α_Mβ₂ Fibγ^390-396A mice with SCA displayed an impressive reduction of reactive oxygen species (ROS) in white blood cells (WBCs), decreased circulating inflammatory cytokines/chemokines, and significantly improved SCA-associated glomerular pathology highlighted by reduced glomerulosclerosis, inflammatory cell infiltration, ischemic lesions, mesangial thickening, mesangial hypercellularity, and glomerular enlargement. In addition, Fibγ^390-396A mice with SCA had improved glomerular protective responses and podocyte/mesangial transcriptional signatures that resulted in reduced albuminuria. Interestingly, the fibrinogen γ^390-396A mutation had a negligible effect on cardiac, lung, and liver functions and pathologies in the context of SCA over a year-long observation period. Taken together, our data support that fibrinogen significantly contributes to WBC-driven inflammation and ROS production, which is a key driver of SCA-associated glomerulopathy, and may represent a novel therapeutic target against irreversible kidney damage in SCA.

Graphical abstract

Figure 1.

Relative ROS production of WBCs is reduced in Fibγ^390-396ASS mice compared with the Fib^WTSS mice. Blood samples from the experimental mice were collected 9 months after BMT and stained with RBCs, WBCs, and platelet-specific antibodies and ROS detection reagent (CM-H2-DCFDA). (A) Relative RBC ROS. (B) Relative WBC ROS. (C) Relative platelets ROS. Fib^WT B6 (n = 12-14), Fibγ^390-396A B6 (n = 11-14), Fib^WT SS (n = 12-14), and Fibγ^390-396A SS (n = 12-14). Similar number of fully chimeric male and female mice were used in the experimental groups. Each symbol in the graph represents an individual mouse. The bars in the graph indicate the mean ± SEM. Statistical significance was determined using 1-way ANOVA followed by the Tukey's multiple comparison test; statistical significance is indicated as **P ≤ .01, ****P ≤ .0001. (A) For RBC ROS, the Dunn's test was used; ***P ≤ .001. ns, not significant.

Figure 2.

Proinflammatory cytokines/chemokines are reduced in Fibγ^390-396ASS mice compared with the Fib^WTSS mice. One year post-BMT, blood samples were collected from the experimental mice. Platelet-poor plasma samples were used for cytokine profiling by Luminex assay. (A) IL-6. (B) TNF-α. (C) KC. Similar number of fully chimeric male and female mice were used in the experimental groups. Each symbol in the graph represents an individual mouse. Fib^WT AA/B6 (n = 11-15), Fibγ^390-396A AA/B6 (n = 8-12), Fib^WT SS (n = 17-22), and Fibγ^390-396A SS (n = 17-22). The bars in the graph indicate the mean ± SEM. Statistical significance was determined using 1-way ANOVA followed by the Tukey's multiple comparison test; statistical significance is indicated as **P ≤ .01, *P ≤ .05.

Figure 3.

Comparison of LV dimensions, systolic function, and mass. (A-B) LVID;d and LVID;s, respectively. (C) LV FS. (D) Echocardiographically derived LV mass is reduced in Fibγ^390-396A SS compared with the Fib^WT SS mice. Echocardiograms were performed on mixed-sex chimeric mice 12 months after BMT. Data are presented as mean ± SEM. Statistical significance was determined using 1-way ANOVA followed by the Tukey's multiple comparison test; statistical significance is indicated as *P ≤ .05, **P ≤ .01, ***P ≤ .001, ****P ≤ .0001.

Figure 4.

Fibγ^390-396Aprotects SCA-associated glomerular pathology. Fully chimeric sickle mice were followed for 1 year posttransplant; kidney samples were fixed in formalin after the mice were euthanized, and stained with H&E and PAS reagent. Representative kidney sections of Fib^WT SS mice (left panel), Fibγ^390-396A SS mice (middle panel) and semiquantitative scoring of histologic features (right panel) showing glomerulosclerosis (A), inflammatory cell infiltration (B), ischemic lesions (C), glomerular mesangial thickening (D), glomerular mesangial hypercellularity (E), and glomerular enlargement (F) are reduced in Fibγ^390-396A SS mice compared with the Fib^WT SS mice. Histopathology scores ranged from 0 to 5, where 0 represents normal kidney morphology; 1, changes in <20% of glomerular area; 2, 21% to 40% of the glomerular area; 3, 41% to 60% of the glomerular area; 4, 61% to 80% of glomerular area; and 5, severe changes in almost all glomeruli. Similar number of fully chimeric male and female mice were used in the experimental groups. The bars in the graph indicate the mean ± SEM. Statistical analyses of the histologic scores were done by Mann-Whitney U test; statistical significance between Fib^WT SS (n = 7) and Fibγ^390-396A SS (n = 8) mice is indicated as **P ≤ .01, *P ≤ .05.

Figure 5.

Urine albumin, urine volume, and blood creatinine are reduced in Fibγ^390-396ASS mice compared with the Fib^WTSS mice. One year after BMT, urine samples were collected for 24 hours from the experimental mice and albumin concentration was measured by the mouse albumin ELISA kit. (A) Urine albumin (normalized by urine creatinine) is reduced in Fibγ^390-396A SS mice compared with the Fib^WT SS mice. (B) Urine volume is decreased in Fibγ^390-396A SS mice compared with the Fib^WT SS mice. One year after BMT, blood samples were collected from the experimental mice. Platelet-poor plasma samples were used for urea nitrogen and creatinine measurement. (C) Blood creatinine is decreased in Fibγ^390-396A SS mice compared with the Fib^WT SS mice. (D) Blood urea nitrogen shows a trend for reduction in Fibγ^390-396A SS mice compared with the Fib^WT SS mice. (E) Urine osmolality is not improved in Fibγ^390-396A SS mice compared with the Fib^WT SS mice. Similar number of fully chimeric male and female mice were used in the experimental groups. Each symbol represents an individual mouse. The bars in the graph indicate the mean ± SEM. Statistical significance was determined using 1-way ANOVA followed by the Tukey's multiple comparison test; statistical significance is indicated as ***P ≤ .001, **P ≤ .01, *P ≤ .05. (D) For blood urea nitrogen, the Dunn test were used.

Figure 6.

Glomerular injury responses to SCA are strongly modified by the Fib γ^390-396domain. RNA-sequencing analyses of glomerular fractions of kidney from the experimental mice 1 year posttransplant reveal chronic injury and loss of normal cellular programming in Fib^WT SS mice compared with the Fibγ^390-396A SS mice. Heatmap data and gene-ontology pathways showing differentially expressed transcripts in Fib^WT SS (n = 3) vs Fibγ^390-396A SS (n = 4) mice glomeruli. Fib^WT B6 (n = 2) and Fibγ^390-396A B6 (n = 3) mice glomeruli were used as control. The most significant enrichments among each of the clusters are indicated by color codes (ToppGene functional enrichments were determined using the complete clusters as shown in supplemental Table 3) and indicate perturbation of the biological processes in the Fib^WT SS and/or the Fibγ^390-396A SS group. Also see supplemental Figure 14 as well as supplemental Table 3 for expanded gene lists associated with each gene cluster. Differentially expressed genes were determined based on Welch ANOVA using a false discovery rate of 0.1 and fold-change >2 for any group comparison.

Figure 7.

Immunofluorescence staining reveals reduced expression of podocyte marker, WT1 in Fib^WTSS compared with the Fibγ^390-396ASS mice. Fully chimeric similar number of male and female mice were followed for 1 year posttransplant and kidney samples were fixed in formalin after the mice were euthanized, and stained with anti-WT1 antibody. The number of WT1⁺ podocytes were counted from 20 glomeruli of each kidney section and average number of WT1⁺ podocytes per glomerulus were calculated. (A) Kidney sections showing similar level of WT1 expression (inside the white dotted circle) in podocytes of glomeruli from Fib^WT B6 (left panel, n = 6) and Fibγ^390-396A B6 (right panel, n = 6) mice. (B) WT1 expression (inside the white dotted circle) is depleted in podocytes of glomeruli of Fib^WT SS (n = 6) mice compared with the Fibγ^390-396A SS (n = 6) mice. The bars in the graph indicate the mean ± SEM. Statistical analyses were done by Mann-Whitney U test and statistical significance are indicated as **P ≤ .01.